Trained on various human-authored corpora, Large Language Models (LLMs) have demonstrated a certain capability of reflecting specific human-like traits (e.g., personality or values) by prompting, benefiting applications like personalized LLMs and social simulations. However, existing methods suffer from the superficial elicitation problem: LLMs can only be steered to mimic shallow and unstable stylistic patterns, failing to embody the desired traits precisely and consistently across diverse tasks like humans. To address this challenge, we propose IROTE, a novel in-context method for stable and transferable trait elicitation. Drawing on psychological theories suggesting that traits are formed through identity-related reflection, our method automatically generates and optimizes a textual self-reflection within prompts, which comprises self-perceived experience, to stimulate LLMs' trait-driven behavior. The optimization is performed by iteratively maximizing an information-theoretic objective that enhances the connections between LLMs' behavior and the target trait, while reducing noisy redundancy in reflection without any fine-tuning, leading to evocative and compact trait reflection. Extensive experiments across three human trait systems manifest that one single IROTE-generated self-reflection can induce LLMs' stable impersonation of the target trait across diverse downstream tasks beyond simple questionnaire answering, consistently outperforming existing strong baselines.

Recently Multimodal Large Language Models (MLLMs) have achieved considerable advancements in vision-language tasks, yet produce potentially harmful or untrustworthy content. Despite substantial work investigating the trustworthiness of language models, MMLMs' capability to act honestly, especially when faced with visually unanswerable questions, remains largely underexplored. This work presents the first systematic assessment of honesty behaviors across various MLLMs. We ground honesty in models' response behaviors to unanswerable visual questions, define four representative types of such questions, and construct MoHoBench, a large-scale MMLM honest benchmark, consisting of 12k+ visual question samples, whose quality is guaranteed by multi-stage filtering and human verification. Using MoHoBench, we benchmarked the honesty of 28 popular MMLMs and conducted a comprehensive analysis. Our findings show that: (1) most models fail to appropriately refuse to answer when necessary, and (2) MMLMs' honesty is not solely a language modeling issue, but is deeply influenced by visual information, necessitating the development of dedicated methods for multimodal honesty alignment. Therefore, we implemented initial alignment methods using supervised and preference learning to improve honesty behavior, providing a foundation for future work on trustworthy MLLMs.

Reinforcement Learning from Human Feedback (RLHF) has emerged as a crucial technique for aligning large language models (LLMs) with human preferences.However, existing RLHF methods face key challenges, including poor sample efficiency, high computational overhead, and slow convergence.Recent studies highlight the importance of data selection in RL, but how to effectively select the most beneficial experiences for RL training remains an open problem.Existing data selection methods for RL rely on heuristic metrics, failing to establish an interpretable connection between data and optimization objectives.To address this problem, we propose InfOES (Influence-based Online Experience Selection), a novel data selection method for RLHF that dynamically estimates the influence of individual training samples on policy optimization.By incorporating data attribution into the policy gradient, InfOES can identify and filter out detrimental samples on the fly, ensuring effective convergence toward alignment objectives.Our approach is compatible with various RL algorithms (e.g., PPO, GRPO, RE-INFORCE++).Extensive experiments demonstrate that InfOES significantly enhances training effectiveness, achieving superior alignment performance with fewer optimization steps.

Bryan Chen Zhengyu Tan, Zhengyuan Liu, Xiaoyuan Yi, Jing Yao, Xing Xie, Nancy F. Chen, Roy Ka-Wei Lee. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Yueqi Xie, Tao Qi, Jingwei Yi, Xiyuan Yang, Ryan Whalen, Junming Huang, Qian Ding, Yu Xie, Xing Xie, Fangzhao Wu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Weihua Zheng, Zhengyuan Liu, Tanmoy Chakraborty, Weiwen Xu, Xiaoxue Gao, Bryan Chen Zhengyu Tan, Bowei Zou, Chang Liu, Yujia Hu, Xing Xie, Xiaoyuan Yi, Jing Yao, Chaojun Wang, Long Li, Rui Liu, Huiyao Liu, Koji Inoue, Ryuichi Sumida, Tatsuya Kawahara, Fan Xu, Lingyu Ye, Wei Tian, Dongjun Kim, Jimin Jung, Jaehyung Seo, Nadya Yuki Wangsajaya, Pham Minh Duc, Ojasva Saxena, Palash Nandi, Xiyan Tao, Wiwik Karlina, Tuan Luong, Keertana Arun Vasan, Roy Ka-Wei Lee, Nancy F. Chen. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Text prompt is the most common way for human-generative AI (GenAI) communication. Though convenient, it is challenging to convey fine-grained and referential intent. One promising solution is to combine text prompts with precise GUI interactions, like brushing and clicking. However, there lacks a formal model to capture synergistic designs between prompts and interactions, hindering their comparison and innovation. To fill this gap, via an iterative and deductive process, we develop the Interaction-Augmented Instruction (IAI) model, a compact entity–relation graph formalizing how the combination of interactions and text prompts enhances human-GenAI communication. With the model, we distill twelve recurring and composable atomic interaction paradigms from prior tools, verifying our model’s capability to facilitate systematic design characterization and comparison. Four usage scenarios further demonstrate the model’s utility in applying, refining, and innovating these paradigms. These results illustrate the IAI model’s descriptive, discriminative, and generative power for shaping future GenAI systems.

AI has been increasingly integrated into screenwriting practice. In refinement, screenwriters expect AI to provide feedback that supports reflection across the internal perspective of characters and the external perspective of the overall story. However, existing AI tools cannot sufficiently coordinate the two perspectives to meet screenwriters’ needs. To address this gap, we present DuoDrama, an AI system that generates feedback to assist screenwriters’ reflection in refinement. To enable DuoDrama, based on performance theories and a formative study with nine professional screenwriters, we design the Experience-Grounded Feedback Generation Workflow for Human Reflection (ExReflect). In ExReflect, an AI agent adopts an experience role to generate experience and then shifts to an evaluation role to generate feedback based on the experience. A study with fourteen professional screenwriters shows that DuoDrama improves feedback quality and alignment and enhances the effectiveness, depth, and richness of reflection. We conclude by discussing broader implications and future directions.

Generative AI has greatly transformed creative work in various domains, such as screenwriting. To understand this transformation, prior research often focused on capturing a snapshot of human-AI co-creation practice at a specific moment, with less attention to how humans mobilize, regulate, and reflect to form the practice gradually. Motivated by Bandura’s theory of human agency, we conducted a two-week study with 19 professional screenwriters to investigate how they embraced AI in their creation process. Our findings revealed that screenwriters not only mindfully planned, foresaw, and responded to AI usage, but, more importantly, through reflections on practice, they developed themselves and human-AI co-creation paradigms, such as cognition, strategies, and workflows. They also expressed various expectations for how future AI should better support their agency. Based on our findings, we conclude this paper with extensive discussion and actionable suggestions to screenwriters, tool developers, and researchers for sustainable human-AI co-creation.

Motivated by the remarkable progress of large language models (LLMs) in objective tasks like mathematics and coding, there is growing interest in their potential to simulate human behavior—a capability with profound implications for transforming social science research and customer-centric business insights. However, LLMs often lack a nuanced understanding of human cognition and behavior, limiting their effectiveness in social simulation and personalized applications. We posit that this limitation stems from a fundamental misalignment: standard LLM pretraining on vast, uncontextualized web data does not capture the continuous, situated context of an individual's decisions, thoughts, and behaviors over time. To bridge this gap, we introduce HumanLLM, a foundation model designed for personalized understanding and simulation of individuals. We first construct the Cognitive Genome Dataset, a large-scale corpus curated from real-world user data on platforms like Reddit, Twitter, Blogger, and Amazon. Through a rigorous, multi-stage pipeline involving data filtering, synthesis, and quality control, we automatically extract over 5.5 million user logs to distill rich profiles, behaviors, and thinking patterns. We then formulate diverse learning tasks and perform supervised fine-tuning to empower the model to predict a wide range of individualized human behaviors, thoughts, and experiences. Comprehensive evaluations demonstrate that HumanLLM achieves superior performance in predicting user actions and inner thoughts, more accurately mimics user writing styles and preferences, and generates more authentic user profiles compared to base models. Furthermore, HumanLLM shows significant gains on out-of-domain social intelligence benchmarks, indicating enhanced generalization. This work paves the way for more human-centric AI systems by advancing research in social simulation, developing personalized companions, enabling marketing intelligence through simulated customer feedback, and powering more realistic user simulation for recommender systems.

Recommender systems suffer from biases that cause the collected feedback to incompletely reveal user preference. While debiasing learning has been extensively studied, they mostly focused on the specialized (calledcounterfactual) test environment simulated by random exposure of items, significantly degrading accuracy in the typical (calledfactual) test environment based on actual user-item interactions. In fact, each test environment highlights the benefit of a different aspect: the counterfactual test emphasizes user satisfaction in the long-terms, while the factual test focuses on predicting subsequent user behaviors on platforms. Therefore, it is desirable to have a model that performs well on both tests rather than only one. In this work, we introduce a new learning framework, calledBias-adaptivePreference distillationLearning (BPL), to gradually uncover user preferences with dual distillation strategies. These distillation strategies are designed to drive high performance in both factual and counterfactual test environments. Employing a specialized form ofteacher-student distillationfrom a biased model, BPL retains accurate preference knowledge aligned with the collected feedback, leading to high performance in the factual test. Furthermore, through self-distillation with reliability filtering, BPL iteratively refines its knowledge throughout the training process. This enables the model to produce more accurate predictions across a broader range of user-item combinations, thereby improving performance in the counterfactual test. Comprehensive experiments validate the effectiveness of BPL in both factual and counterfactual tests.

The application scope of Large Language Models (LLMs) continues to expand, leading to increasing interest in personalized LLMs that align with human values.However, aligning these models with individual values raises significant safety concerns, as certain values may correlate with harmful information.In this paper, we identify specific safety risks associated with value-aligned LLMs and investigate the psychological principles behind these challenges.Our findings reveal two key insights.(1) Value-aligned LLMs are more prone to harmful behavior compared to non-fine-tuned models and exhibit slightly higher risks in traditional safety evaluations than other fine-tuned models.(2) These safety issues arise because value-aligned LLMs genuinely generate text according to the aligned values, which can amplify harmful outcomes.Using a dataset with detailed safety categories, we find significant correlations between value alignment and safety risks, supported by psychological hypotheses.This study offers insights into the "black box" of value alignment and proposes in-context alignment methods to enhance the safety of value-aligned LLMs. 1

Yuhong Dai, Jianxun Lian, Yitian Huang, Wei Zhang, Mingyang Zhou, Mingqi Wu, Xing Xie, Hao Liao. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

As Large Language Models (LLMs) advance, aligning them with human values is critical for their responsible development.Value principles serve as the foundation for clarifying alignment goals.Multiple sets of value principles have been proposed, such as HHH (helpful, honest, harmless) and instructions for data synthesis in reinforcement learning from AI feedback (RLAIF).However, most of them are heuristically crafted, without consideration of three primary challenges in practical LLM alignment: 1) Comprehensiveness to deal with diverse and even unforeseen scenarios in which LLMs could be applied; 2) Precision to provide LLMs with clear and actionable guidance in specific scenarios; and 3) Compatability to avoid internal contracts between principles.In this paper, we formalize quantitative metrics to evaluate value principles along the three desirable properties.Building on these metrics, we propose the Hierarchical Value Principle framework (HiVaP) 1 , which constructs a hierarchical principle set and retrieves principles tailored to each scenario in a cascading way, addressing above challenges.Experimental results validate that the three metrics capture the effectiveness of value principles for LLM alignment, and our HiVaP framework that enhances these metrics leads to superior alignment.

Ziyu Chen, Junfei Sun, Chenxi Li, Tuan Dung Nguyen, Jing Yao, Xiaoyuan Yi, Xing Xie, Chenhao Tan, Lexing Xie. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

*Warning: Contains harmful model outputs.* Despite significant advancements, the propensity of Large Language Models (LLMs) to generate harmful and unethical content poses critical challenges. Measuring value alignment of LLMs becomes crucial for their regulation and responsible deployment. Although numerous benchmarks have been constructed to assess social bias, toxicity, and ethical issues in LLMs, those static benchmarks suffer from *evaluation chronoeffect*, in which, as models rapidly evolve, existing benchmarks may leak into training data or become saturated, *overestimating* ever-developing LLMs. To tackle this problem, we propose GETA, a novel *generative evolving testing* approach based on adaptive testing methods in measurement theory. Unlike traditional adaptive testing methods that rely on a static test item pool, GETA probes the underlying moral boundaries of LLMs by dynamically generating test items tailored to model capability. GETA co-evolves with LLMs by learning a joint distribution of item difficulty and model value conformity, thus effectively addressing evaluation chronoeffect. We evaluated various popular LLMs with GETA and demonstrated that 1) GETA can dynamically create difficulty-tailored test items and 2) GETA's evaluation results are more consistent with models' performance on unseen OOD and i.i.d. items, laying the groundwork for future evaluation paradigms.

The integration of large language models (LLMs) with external content has enabled applications such as Microsoft Copilot but also introduced vulnerabilities to indirect prompt injection attacks. In these attacks, malicious instructions embedded within external content can manipulate LLM outputs, causing deviations from user expectations. To address this critical yet under-explored issue, we introduce the first benchmark for bindirect prompt injection attacks, named BIPIA, to assess the risk of such vulnerabilities. Using BIPIA, we evaluate existing LLMs and find them universally vulnerable. Our analysis identifies two key factors contributing to their success: LLMs' inability to distinguish between informational context and actionable instructions, and their lack of awareness in avoiding the execution of instructions within external content. Based on these findings, we propose two novel defense mechanisms -- boundary awareness and explicit reminder -- to address these vulnerabilities in both black-box and white-box settings. Extensive experiments demonstrate that our black-box defense provides substantial mitigation, while our white-box defense reduces the attack success rate to near-zero levels, all while preserving the output quality of LLMs. We hope this work inspires further research into securing LLM applications and fostering their safe and reliable use. Our code is available at https://github.com/microsoft/BIPIA.

Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities but raise significant privacy concerns due to their abilities to infer sensitive personal information from images with high precision. While current LVLMs are relatively well aligned to protect universal privacy, e.g., credit card data, we argue that privacy is inherently personalized and context-dependent. This work pivots towards a novel task: can LVLMs achieve Inference-Time Personalized Privacy Protection (ITP3), allowing users to dynamically specify privacy boundaries through language specifications? To this end, we present SPY-Bench, the first systematic assessment of ITP3 ability, which comprises (1) 32,700 unique samples with image-question pairs and personalized privacy instructions across 67 categories and 24 real-world scenarios, and (2) novel metrics grounded in user specifications and context awareness. Benchmarking the ITP3 ability of 21 SOTA LVLMs, we reveal that: (i) most models, even the top-performing o4-mini, perform poorly, with only ~24% compliance accuracy; (ii) they show quite limited contextual privacy understanding capability. Therefore, we implemented initial ITP3 alignment methods, including a novel Noise Contrastive Alignment variant which achieves 96.88% accuracy while maintaining reasonable general performance. These results mark an initial step towards the ethical deployment of more controllable LVLMs. Code and data are at https://github.com/achernarwang/specify-privacy-yourself.

Lei Wang, Jianxun Lian, Yi Huang, Yanqi Dai, Haoxuan Li, Xu Chen, Xing Xie, Ji-Rong Wen. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

As large language models (LLMs) become increasingly integrated into applications serving users across diverse cultures, communities, and demographics, it is critical to align LLMs with pluralistic human values beyond average principles (e.g., HHH). In psychological and social value theories such as Schwartz’s Value Theory, pluralistic values are represented by multiple value dimensions paired with various priorities. However, existing methods encounter two challenges when aligning with such fine-grained value objectives: 1) they often treat multiple values as independent and equally important, ignoring their interdependence and relative priorities (value complexity); 2) they struggle to precisely control nuanced value priorities, especially those underrepresented ones (value steerability). To handle these challenges, we propose COUPLE, a COUnterfactual reasoning framework for PLuralistic valuE alignment. It introduces a structural causal model (SCM) to feature complex interdependency and prioritization among features, as well as the causal relationship between high-level value dimensions and behaviors. Moreover, it applies counterfactual reasoning to generate outputs aligned with any desired value objectives. Benefitting from explicit causal modeling, COUPLE also provides better interpretability. We evaluate COUPLE on two datasets with different value systems and demonstrate that COUPLE advances other baselines across diverse types of value objectives. Our code is available at https://github.com/microsoft/COUPLE.

Large language models (LLMs) have shown impressive capabilities across tasks such as mathematics, coding, and reasoning, yet their learning ability, which is crucial for adapting to dynamic environments and acquiring new knowledge, remains underexplored. In this work, we address this gap by introducing a framework inspired by cognitive psychology and education. Specifically, we decompose general learning ability into three distinct, complementary dimensions: *Learning from Instructor* (acquiring knowledge via explicit guidance), *Learning from Concept* (internalizing abstract structures and generalizing to new contexts), and *Learning from Experience* (adapting through accumulated exploration and feedback). We conduct a comprehensive empirical study across the three learning dimensions and identify several insightful findings, such as (i) interaction improves learning; (ii) conceptual understanding is scale-emergent and benefits larger models; and (iii) LLMs are effective few-shot learners but not many-shot learners. Based on our framework and empirical findings, we introduce a benchmark that provides a unified and realistic evaluation of LLMs' general learning abilities across three learning cognition dimensions. It enables diagnostic insights and supports evaluation and development of more adaptive and human-like models.

The spread of fake news harms individuals and presents a critical social challenge that must be addressed. Although numerous algorithmic and insightful features have been developed to detect fake news, many of these features can be manipulated with style-conversion attacks, especially with the emergence of advanced language models, making it more difficult to differentiate from genuine news. This study proposes adversarial style augmentation, AdStyle, designed to train a fake news detector that remains robust against various style-conversion attacks. The primary mechanism involves the strategic use of LLMs to automatically generate a diverse and coherent array of style-conversion attack prompts, enhancing the generation of particularly challenging prompts for the detector. Experiments indicate that our augmentation strategy significantly improves robustness and detection performance when evaluated on fake news benchmark datasets.

Foundation models are usually pre-trained on large-scale datasets and then adapted to different downstream tasks through tuning. This pre-training and then fine-tuning paradigm has become a standard practice in deep learning. However, the large-scale pre-training datasets, often inaccessible or too expensive to handle, can contain label noise that may adversely affect the generalization of the model and pose unexpected risks. This paper stands out as the first work to comprehensively understand and analyze the nature of noise in pre-training datasets and then effectively mitigate its impacts on downstream tasks. Specifically, through extensive experiments of fully-supervised and image-text contrastive pre-training on synthetic noisy ImageNet-1 K, YFCC15 M, and CC12 M datasets, we demonstrate that, while slight noise in pre-training can benefit in-domain (ID) performance, where the training and testing data share a similar distribution, it always deteriorates out-of-domain (OOD) performance, where training and testing distributions are significantly different. These observations are agnostic to scales of pre-training datasets, pre-training noise types, model architectures, pre-training objectives, downstream tuning methods, and downstream applications. We empirically ascertain that the reason behind this is that the pre-training noise shapes the feature space differently. We then propose a tuning method (NMTune) to affine the feature space to mitigate the malignant effect of noise and improve generalization, which is applicable in both parameter-efficient and black-box tuning manners, considering one may not be able to access or fully fine-tune the pre-trained models. We additionally conduct extensive experiments on popular vision and language models, including APIs, which are supervised and self-supervised pre-trained on realistic noisy data for evaluation. Our analysis and results demonstrate the importance of this novel and fundamental research direction, which we term as Noisy Model Transfer Learning.

Explainable recommender systems ( ERS ) aim to enhance users’ trust in the systems by offering personalized recommendations with transparent explanations. This transparency provides users with a clear understanding of the rationale behind the recommendations, fostering a sense of confidence and reliability in the system’s outputs. Generally, the explanations are presented in a familiar and intuitive way, which is in the form of natural language, thus enhancing their accessibility to users. Recently, there has been an increasing focus on leveraging reviews as a valuable source of rich information in both modeling user-item preferences and generating textual interpretations, which can be performed simultaneously in a multi-task framework. Despite the progress made in these review-based recommendation systems, the integration of implicit feedback derived from user-item interactions and user-written text reviews has yet to be fully explored. To fill this gap, we propose a model named SERMON (A s pect-enhanced E xplainable R ecommendation with M ulti-modal C o ntrast Lear n ing). Our model explores the application of multimodal contrastive learning to facilitate reciprocal learning across two modalities, thereby enhancing the modeling of user preferences. Moreover, our model incorporates the aspect information extracted from the review, which provides two significant enhancements to our tasks. Firstly, the quality of the generated explanations is improved by incorporating the aspect characteristics into the explanations generated by a pre-trained model with controlled textual generation ability. Secondly, the commonly used user-item interactions are transformed into user-item-aspect interactions, which we refer to as interaction triple, resulting in a more nuanced representation of user preference. To validate the effectiveness of our model, we conduct extensive experiments on three real-world datasets. The experimental results show that our model outperforms state-of-the-art baselines, with a 2.0% improvement in prediction accuracy and a substantial 24.5% enhancement in explanation quality for the TripAdvisor dataset.

Meta learning has been recognized as an effective remedy for solving the cold-start problem in the recommendation domain. Existing models aim to learn how to generalize from the user behaviors in the training set to testing set. However, in the cold start settings, with only a small number of training samples, the testing distribution may easily deviate from the training one, which may invalidate the learned generalization patterns, and lower the recommendation performance. For alleviating this problem, in this paper, we propose a robust meta recommender framework to address the distribution shift problem. In specific, we argue that the distribution shift may exist on both the user- and interaction-levels, and in order to mitigate them simultaneously, we design a novel distributionally robust model by hierarchically reweighing the training samples. Different sample weights correspond to different training distributions, and we minimize the largest loss induced by the sample weights in a simplex, which essentially optimizes the upper bound of the testing loss. In addition, we analyze our framework on the convergence rates and generalization error bound to provide more theoretical insights. Empirically, we conduct extensive experiments based on different meta recommender models and real-world datasets to verify the generality and effectiveness of our framework.

Recommender models capture ever-changing user preferences by training with in-domain user behavior data. These models are typically lightweight, facilitating real-time and large-scale online services. However, these models often falter when tasked with providing more sophisticated functionalities, such as offering explanations or engaging in conversations. Recently, large language models (LLMs) have emerged as a significant advancement towards artificial general intelligence, demonstrating impressive capabilities in instruction comprehension, reasoning, and human interaction. Unfortunately, LLMs lack the understanding of domain-specific item catalogs and behavioral patterns, especially in areas that deviate from general world knowledge, such as online e-commerce. This limitation makes them unsuitable to function as recommender models directly. In this article, we bridge the gap between recommender models and LLMs, combining their respective strengths to create an interactive recommender system. We present an efficient framework, termed as InteRecAgent , which utilizes LLMs as the brain and recommender models as instrumental tools. We first outline a minimal set of essential tools required to transform LLMs into InteRecAgent. To overcome specific challenges associated with LLM-based agents for recommender systems, we enhance three core components, covering memory mechanism, task planning, and tool learning abilities. The InteRecAgent empowers traditional recommender systems, like ID-based matrix factorization models, to evolve into versatile and interactive systems with a natural language interface through the integration of LLMs. Experimental results derived from three public datasets demonstrate that the InteRecAgent delivers strong performance as a conversational recommender system, surpassing general LLMs such as GPT-4.

Explainability is critical for recommender systems to ensure good user experience and facilitate designers to debug. However, generating explanations in recommender systems usually requires large efforts due to the dependency on additional data and case-by-case model design. One possible solution to these challenges is reasoning with logic rules, whose validity or confidence can automatically indicate high-quality explanations and formats are general. However, pioneer methods can be hardly applied in recommendation due to the high sparsity of interaction data, which raises the difficulty in accurately computing the rule validity, and the specific ranking-oriented task. To bridge this gap, we propose a general framework for Reco mmendation with lo gic r ule reasoning ( Recolor ) that satisfies three desirable properties. First, we explicitly estimate the rule validity to ensure well-grounded decisions, where a fuzzy logic validity module is designed for accurate estimation on highly sparse recommendation data. Second, we ensure the generality for both the types of input data and model architectures by designing a neural logic generation module, which decouples the user–item representation learning from the rule construction. Third, we integrate the two above-mentioned modules with a ranking-oriented BPR loss and achieve a unified optimization of explainability and accuracy. For any given neural recommendation model, our proposed logic rule reasoning framework can upgrade it to a self-explainable version. Numerical experiments and user studies on four public recommendation datasets with different levels of sparsity demonstrate that our framework shows high-validity rule explanations, generality in architecture and data, and high recommendation accuracy.

In recent years, recommender systems have achieved remarkable performance by using ensembles of heterogeneous models. However, this approach is costly due to the resources and inference latency proportional to the number of models, creating a bottleneck for production. Our work aims at transfering the ensemble knowledge of heterogeneous teachers to a lightweight student model using knowledge distillation (KD), reducing inference costs while maintaining high accuracy. We find that the efficacy of distillation decreases when transferring knowledge from heterogeneous teachers. To address this, we propose a new KD framework, named HetComp, that guides the student model by transferring easy-to-hard sequences of knowledge generated from teachers’ trajectories. HetComp uses dynamic knowledge construction to provide progressively difficult ranking knowledge and adaptive knowledge transfer to gradually transfer finer-grained ranking information. Although HetComp improves accuracy, it exacerbates popularity bias, resulting in a high popularity lift. To mitigate this issue, we introduce two strategies that leverage models’ disagreement knowledge (i.e., dissensus) for heterogeneous comparison. Our experiments demonstrate that HetComp significantly enhances distillation quality and the student model’s generalization capabilities. Furthermore, we provide extensive experimental results supporting the effectiveness of our dissensus-based debiasing techniques in mitigating the popularity lift caused by HetComp.

Retrieval models aim at selecting a small set of item candidates which match the preference of a given user. They play a vital role in large-scale recommender systems since subsequent models such as rankers highly depend on the quality of item candidates. However, most existing retrieval models employ a single-round inference paradigm, which may not adequately capture the dynamic nature of user preferences and stuck in one area in the item space. In this paper, we propose Ada-Retrieval, an adaptive multi-round retrieval paradigm for recommender systems that iteratively refines user representations to better capture potential candidates in the full item space. Ada-Retrieval comprises two key modules: the item representation adapter and the user representation adapter, designed to inject context information into items' and users' representations. The framework maintains a model-agnostic design, allowing seamless integration with various backbone models such as RNNs or Transformers. We perform experiments on three widely used public datasets, incorporating five powerful sequential recommenders as backbone models. Our results demonstrate that Ada-Retrieval significantly enhances the performance of various base models, with consistent improvements observed across different datasets. Our code and data are publicly available at: https://github.com/ll0ruc/Ada-Retrieval.

Wensheng Lu, Jianxun Lian, Wei Zhang, Guanghua Li, Mingyang Zhou, Hao Liao, Xing Xie. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Zhuohao Yu, Chang Gao, Wenjin Yao, Yidong Wang, Wei Ye, Jindong Wang, Xing Xie, Yue Zhang, Shikun Zhang. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Pre-training on large-scale datasets and then fine-tuning on downstream tasks have become a standard practice in deep learning. However, pre-training data often contain label noise that may adversely affect the generalization of the model. This paper aims to understand the nature of noise in pre-training datasets and to mitigate its impact on downstream tasks. More specifically, through extensive experiments of supervised pre-training models on synthetic noisy ImageNet-1K and YFCC15M datasets, we demonstrate that while slight noise in pre-training can benefit in-domain (ID) transfer performance, where the training and testing data share the same distribution, it always deteriorates out-of-domain (OOD) performance, where training and testing data distribution are different. We empirically verify that the reason behind is noise in pre-training shapes the feature space differently. We then propose a light-weight black-box tuning method (NMTune) to affine the feature space to mitigate the malignant effect of noise and improve generalization on both ID and OOD tasks, considering one may not be able to fully fine-tune or even access the pre-trained models. We conduct practical experiments on popular vision and language models that are pre-trained on noisy data for evaluation of our approach. Our analysis and results show the importance of this interesting and novel research direction, which we term Noisy Model Learning.

Large Language Models (LLMs) have made unprecedented breakthroughs, yet their increasing integration into everyday life might raise societal risks due to generated unethical content. Despite extensive study on specific issues like bias, the intrinsic values of LLMs remain largely unexplored from a moral philosophy perspective. This work delves into ethical values utilizing Moral Foundation Theory. Moving beyond conventional discriminative evaluations with poor reliability, we propose DeNEVIL, a novel prompt generation algorithm tailored to dynamically exploit LLMs’ value vulnerabilities and elicit the violation of ethics in a generative manner, revealing their underlying value inclinations. On such a basis, we construct MoralPrompt, a high-quality dataset comprising 2,397 prompts covering 500+ value principles, and then benchmark the intrinsic values across a spectrum of LLMs. We discovered that most models are essentially misaligned, necessitating further ethical value alignment. In response, we develop VILMO, an in-context alignment method that substantially enhances the value compliance of LLM outputs by learning to generate appropriate value instructions, outperforming existing competitors. Our methods are suitable for black-box and open-source models, offering a promising initial step in studying the ethical values of LLMs.

Instruction tuning large language models (LLMs) remains a challenging task, owing to the complexity of hyperparameter selection and the difficulty involved in evaluating the tuned models. To determine the optimal hyperparameters, an automatic, robust, and reliable evaluation benchmark is essential. However, establishing such a benchmark is not a trivial task due to the challenges associated with evaluation accuracy and privacy protection. In response to these challenges, we introduce a judge large language model, named PandaLM, which is trained to distinguish the superior model given several LLMs. PandaLM's focus extends beyond just the objective correctness of responses, which is the main focus of traditional evaluation datasets. It addresses vital subjective factors such as relative conciseness, clarity, adherence to instructions, comprehensiveness, and formality. To ensure the reliability of PandaLM, we collect a diverse human-annotated test dataset, where all contexts are generated by humans and labels are aligned with human preferences. Our findings reveal that PandaLM-7B offers a performance comparable to both GPT-3.5 and GPT-4. Impressively, PandaLM-70B surpasses their performance. PandaLM enables the evaluation of LLM to be fairer but with less cost, evidenced by significant improvements achieved by models tuned through PandaLM compared to their counterparts trained with default Alpaca's hyperparameters. In addition, PandaLM does not depend on API-based evaluations, thus avoiding potential data leakage.

Large Language Models (LLMs) exhibit emerging in-context learning abilities through prompt engineering. The recent progress in large-scale generative models has further expanded their use in real-world language applications. However, the critical challenge of improving the generalizability and factuality of LLMs in natural language understanding and question answering remains under-explored. While previous in-context learning research has focused on enhancing models to adhere to users' specific instructions and quality expectations, and to avoid undesired outputs, little to no work has explored the use of task-specific fine-tuned Language Models (SLMs) to improve LLMs' in-context learning during the inference stage. Our primary contribution is the establishment of a simple yet effective framework that enhances the reliability of LLMs as it: 1) generalizes out-of-distribution data, 2) elucidates how LLMs benefit from discriminative models, and 3) minimizes hallucinations in generative tasks. Using our proposed plug-in method, enhanced versions of Llama 2 and ChatGPT surpass their original versions regarding generalizability and factuality. We offer a comprehensive suite of resources, including 16 curated datasets, prompts, model checkpoints, and LLM outputs across 9 distinct tasks. Our empirical analysis sheds light on the advantages of incorporating discriminative models into LLMs and highlights the potential of our methodology in fostering more reliable LLMs.

Machine learning has demonstrated remarkable performance over finite datasets, yet whether the scores over the fixed benchmarks can sufficiently indicate the model’s performance in the real world is still in discussion. In reality, an ideal robust model will probably behave similarly to the oracle (e.g., the human users), thus a good evaluation protocol is probably to evaluate the models’ behaviors in comparison to the oracle. In this paper, we introduce a new robustness measurement that directly measures the image classification model’s performance compared with a surrogate oracle (i.e., a zoo of foundation models). Besides, we design a simple method that can accomplish the evaluation beyond the scope of the benchmarks. Our method extends the image datasets with new samples that are sufficiently perturbed to be distinct from the ones in the original sets, but are still bounded within the same image-label structure the original test image represents, constrained by a zoo of foundation models pretrained with a large amount of samples. As a result, our new method will offer us a new way to evaluate the models’ robustness performance, free of limitations of fixed benchmarks or constrained perturbations, although scoped by the power of the oracle. In addition to the evaluation results, we also leverage our generated data to understand the behaviors of the model and our new evaluation strategies.

Large language models (LLMs) have achieved remarkable performance in various evaluation benchmarks. However, concerns are raised about potential data contamination in their considerable volume of training corpus. Moreover, the static nature and fixed complexity of current benchmarks may inadequately gauge the advancing capabilities of LLMs. In this paper, we introduce DyVal, a general and flexible protocol for dynamic evaluation of LLMs. Based on our framework, we build graph-informed DyVal by leveraging the structural advantage of directed acyclic graphs to dynamically generate evaluation samples with controllable complexities. DyVal generates challenging evaluation sets on reasoning tasks including mathematics, logical reasoning, and algorithm problems. We evaluate various LLMs ranging from Flan-T5-large to GPT-3.5-Turbo and GPT-4. Experiments show that LLMs perform worse in DyVal-generated evaluation samples with different complexities, highlighting the significance of dynamic evaluation. We also analyze the failure cases and results of different prompting methods. Moreover, DyVal-generated samples are not only evaluation sets, but also helpful data for fine-tuning to improve the performance of LLMs on existing benchmarks. We hope that DyVal can shed light on future evaluation research of LLMs. Code is available at: https://github.com/microsoft/promptbench.

Weakly supervised learning generally faces challenges in applicability to various scenarios with diverse weak supervision and in scalability due to the complexity of existing algorithms, thereby hindering the practical deployment. This paper introduces a general framework for learning from weak supervision (GLWS) with a novel algorithm. Central to GLWS is an Expectation-Maximization (EM) formulation, adeptly accommodating various weak supervision sources, including instance partial labels, aggregate statistics, pairwise observations, and unlabeled data. We further present an advanced algorithm that significantly simplifies the EM computational demands using a Non-deterministic Finite Automaton (NFA) along with a forward-backward algorithm, which effectively reduces time complexity from quadratic or factorial often required in existing solutions to linear scale. The problem of learning from arbitrary weak supervision is therefore converted to the NFA modeling of them. GLWS not only enhances the scalability of machine learning models but also demonstrates superior performance and versatility across 11 weak supervision scenarios. We hope our work paves the way for further advancements and practical deployment in this field.

Large language models (LLMs) have gained considerable attention for their excellent natural language processing capabilities. Nonetheless, these LLMs present many challenges, particularly in the realm of trustworthiness. This paper introduces TrustLLM, a comprehensive study of trustworthiness in LLMs, including principles for different dimensions of trustworthiness, established benchmark, evaluation, and analysis of trustworthiness for mainstream LLMs, and discussion of open challenges and future directions. Specifically, we first propose a set of principles for trustworthy LLMs that span eight different dimensions. Based on these principles, we further establish a benchmark across six dimensions including truthfulness, safety, fairness, robustness, privacy, and machine ethics. We then present a study evaluating 16 mainstream LLMs in TrustLLM, consisting of over 30 datasets. Our findings firstly show that in general trustworthiness and capability (i.e., functional effectiveness) are positively related. Secondly, our observations reveal that proprietary LLMs generally outperform most open-source counterparts in terms of trustworthiness, raising concerns about the potential risks of widely accessible open-source LLMs. However, a few open-source LLMs come very close to proprietary ones, suggesting that open-source models can achieve high levels of trustworthiness without additional mechanisms like moderator, offering valuable insights for developers in this field. Thirdly, it is important to note that some LLMs may be overly calibrated towards exhibiting trustworthiness, to the extent that they compromise their utility by mistakenly treating benign prompts as harmful and consequently not responding. Besides these observations, we’ve uncovered key insights into the multifaceted trustworthiness in LLMs. We emphasize the importance of ensuring transparency not only in the models themselves but also in the technologies that underpin trustworthiness. We advocate that the establishment of an AI alliance between industry, academia, the open-source community to foster collaboration is imperative to advance the trustworthiness of LLMs.

Emotion significantly impacts our daily behaviors and interactions. While recent generative AI models, such as large language models, have shown impressive performance in various tasks, it remains unclear whether they truly comprehend emotions and why. This paper aims to address this gap by incorporating psychological theories to gain a holistic understanding of emotions in generative AI models. Specifically, we propose three approaches: 1) EmotionPrompt to enhance AI model performance, 2) EmotionAttack to impair AI model performance, and 3) EmotionDecode to explain the effects of emotional stimuli, both benign and malignant. Through extensive experiments involving language and multi-modal models on semantic understanding, logical reasoning, and generation tasks, we demonstrate that both textual and visual EmotionPrompt can boost the performance of AI models while EmotionAttack can hinder it. More importantly, EmotionDecode reveals that AI models can comprehend emotional stimuli akin to the mechanism of dopamine in the human brain. Our work heralds a novel avenue for exploring psychology to enhance our understanding of generative AI models, thus boosting the research and development of human-AI collaboration and mitigating potential risks.

Large language models (LLMs) have been widely used as agents to complete different tasks, such as personal assistance or event planning. Although most of the work has focused on cooperation and collaboration between agents, little work explores competition, another important mechanism that promotes the development of society and economy. In this paper, we seek to examine the competition dynamics in LLM-based agents. We first propose a general framework for studying the competition between agents. Then, we implement a practical competitive environment using GPT-4 to simulate a virtual town with two types of agents, including restaurant agents and customer agents. Specifically, the restaurant agents compete with each other to attract more customers, where competition encourages them to transform, such as cultivating new operating strategies. Simulation experiments reveal several interesting findings at the micro and macro levels, which align well with existing market and sociological theories. We hope that the framework and environment can be a promising testbed to study the competition that fosters understanding of society. Code is available at: https://github.com/microsoft/competeai.

Evaluation of large language models (LLMs) has raised great concerns in the community due to the issue of data contamination. Existing work designed evaluation protocols using well-defined algorithms for specific tasks, which cannot be easily extended to diverse scenarios. Moreover, current evaluation benchmarks can only provide the overall benchmark results and cannot support a fine-grained and multifaceted analysis of LLMs’ abilities. In this paper, we propose meta probing agents (MPA), a general dynamic evaluation protocol inspired by psychometrics to evaluate LLMs. MPA designs the probing and judging agents to automatically transform an original evaluation problem into a new one following psychometric theory on three basic cognitive abilities: language understanding, problem solving, and domain knowledge. These basic abilities are also dynamically configurable, allowing multifaceted analysis. We conducted extensive evaluations using MPA and found that most LLMs achieve poorer performance, indicating room for improvement. Our multifaceted analysis demonstrated the strong correlation between the basic abilities and an implicit Mattew effect on model size, i.e., larger models possess stronger correlations of the abilities. MPA can also be used as a data augmentation approach to enhance LLMs. Code is available at: https://github.com/microsoft/promptbench.

Recent advancements in fully-supervised change detection (CD) have been notable, yet applying CD effectively in unlabeled scenarios remains a challenge. Our study focuses on domain adaptation of change detection (DACD), transferring knowledge from a labeled dataset (source domain) to an unlabeled dataset (target domain). However, we’ve found that this transfer often leads to reduced discrimination between change and non-change classes because of cross-domain discrepancy, adversely affecting CD performance in the target domain. Aiming at it, we propose a source discriminative learning (SDL) method to learn more discriminative interclass feature representations in the source domain, which can improve the transferability of change knowledge across different CD domains. To evaluate its effectiveness, we established two cross-domain CD scenarios using well-known CD datasets and applied our SDL method to them. The experimental results affirm that our method effectively boosts DACD performance.

Recommender systems are widely used in online services, with embedding-based models being particularly popular due to their expressiveness in representing complex signals. However, these models often function as a black box, making them less transparent and reliable for both users and developers. Recently, large language models (LLMs) have demonstrated remarkable intelligence in understanding, reasoning, and instruction following. This paper presents the initial exploration of using LLMs as surrogate models to explaining black-box recommender models. The primary concept involves training LLMs to comprehend and emulate the behavior of target recommender models. By leveraging LLMs' own extensive world knowledge and multi-step reasoning abilities, these aligned LLMs can serve as advanced surrogates, capable of reasoning about observations. Moreover, employing natural language as an interface allows for the creation of customizable explanations that can be adapted to individual user preferences. To facilitate an effective alignment, we introduce three methods: behavior alignment, intention alignment, and hybrid alignment. Behavior alignment operates in the language space, representing user preferences and item information as text to mimic the target model's behavior; intention alignment works in the latent space of the recommendation model, using user and item representations to understand the model's behavior; hybrid alignment combines both language and latent spaces. Comprehensive experiments conducted on three public datasets show that our approach yields promising results in understanding and mimicking target models, producing high-quality, high-fidelity, and distinct explanations. Our code is available at https://github.com/microsoft/RecAI.

Recent advancements in diffusion models trained on large-scale data have enabled the generation of indistinguishable human-level images, yet they often produce harmful content misaligned with human values, e.g., social bias, and offensive content. Despite extensive research on Large Language Models (LLMs), the challenge of Text-to-Image (T2I) model alignment remains largely unexplored. Addressing this problem, we propose LiVO (Lightweight Value Optimization), a novel lightweight method for aligning T2I models with human values. LiVO only optimizes a plug-and-play value encoder to integrate a specified value principle with the input prompt, allowing the control of generated images over both semantics and values. Specifically, we design a diffusion model-tailored preference optimization loss, which theoretically approximates the Bradley-Terry model used in LLM alignment but provides a more flexible trade-off between image quality and value conformity. To optimize the value encoder, we also develop a framework to automatically construct a text-image preference dataset of 86k (prompt, aligned image, violating image, value principle) samples. Without updating most model parameters and through adaptive value selection from the input prompt, LiVO significantly reduces harmful outputs and achieves faster convergence, surpassing several strong baselines and taking an initial step towards ethically aligned T2I models. Warning: This paper involves descriptions and images depicting discriminatory, pornographic, bloody, and horrific scenes.

Jing Yao, Xiaoyuan Yi, Yifan Gong, Xiting Wang, Xing Xie. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Learning with reduced labeling standards, such as noisy label, partial label, and supplementary unlabeled data, which we generically refer to as imprecise label, is a commonplace challenge in machine learning tasks. Previous methods tend to propose specific designs for every emerging imprecise label configuration, which is usually unsustainable when multiple configurations of imprecision coexist. In this paper, we introduce imprecise label learning (ILL), a framework for the unification of learning with various imprecise label configurations. ILL leverages expectation-maximization (EM) for modeling the imprecise label information, treating the precise labels as latent variables. Instead of approximating the correct labels for training, it considers the entire distribution of all possible labeling entailed by the imprecise information. We demonstrate that ILL can seamlessly adapt to partial label learning, semi-supervised learning, noisy label learning, and, more importantly, a mixture of these settings, with closed-form learning objectives derived from the unified EM modeling. Notably, ILL surpasses the existing specified techniques for handling imprecise labels, marking the first practical and unified framework with robust and effective performance across various challenging settings. We hope our work will inspire further research on this topic, unleashing the full potential of ILL in wider scenarios where precise labels are expensive and complicated to obtain.

Large language models (LLMs) have been observed to exhibit bias towards certain cultures due to the predominance of training data obtained from English corpora. Considering that multilingual cultural data is often expensive to procure, existing methodologies address this challenge through prompt engineering or culture-specific pre-training. However, these strategies may neglect the knowledge deficiency of low-resource cultures and necessitate substantial computing resources. In this paper, we propose CultureLLM, a cost-effective solution to integrate cultural differences into LLMs. CultureLLM employs the World Value Survey (WVS) as seed data and generates semantically equivalent training data through the proposed semantic data augmentation. Utilizing only $50$ seed samples from WVS with augmented data, we fine-tune culture-specific LLMs as well as a unified model (CultureLLM-One) for $9$ cultures, encompassing both rich and low-resource languages. Extensive experiments conducted on $60$ culture-related datasets reveal that CultureLLM significantly surpasses various counterparts such as GPT-3.5 (by $8.1$\%) and Gemini Pro (by $9.5$\%), demonstrating performance comparable to or exceeding that of GPT-4. Our human study indicates that the generated samples maintain semantic equivalence to the original samples, offering an effective solution for LLMs augmentation. Code is released at https://github.com/Scarelette/CultureLLM.

Cultural bias is pervasive in many large language models (LLMs), largely due to the deficiency of data representative of different cultures. Typically, cultural datasets and benchmarks are constructed either by extracting subsets of existing datasets or by aggregating from platforms such as Wikipedia and social media. However, these approaches are highly dependent on real-world data and human annotations, making them costly and difficult to scale. Inspired by cognitive theories on social communication, this paper introduces CulturePark, an LLM-powered multi-agent communication framework for cultural data collection. CulturePark simulates cross-cultural human communication with LLM-based agents playing roles in different cultures. It generates high-quality cross-cultural dialogues encapsulating human beliefs, norms, and customs. Using CulturePark, we generated 41,000 cultural samples to fine-tune eight culture-specific LLMs. We evaluated these models across three downstream tasks: content moderation, cultural alignment, and cultural education. Results show that for content moderation, our GPT-3.5-based models either match or outperform GPT-4 on $41$ datasets. Regarding cultural alignment, our models surpass GPT-4 on Hofstede's VSM 13 framework. Furthermore, for cultural education of human participants, our models demonstrate superior outcomes in both learning efficacy and user experience compared to GPT-4. CulturePark proves an important step in addressing cultural bias and advancing the democratization of AI, highlighting the critical role of culturally inclusive data in model training. Code is released at https://github.com/Scarelette/CulturePark.

Large language models (LLMs) have achieved unprecedented performances in various applications, yet evaluating them is still challenging. Existing benchmarks are either manually constructed or are automatic, but lack the ability to evaluate the thought process of LLMs with arbitrary complexity. We contend that utilizing existing relational databases based on the entity-relationship (ER) model is a promising approach for constructing benchmarks as they contain structured knowledge that can be used to question LLMs. Unlike knowledge graphs, which are also used to evaluate LLMs, relational databases have integrity constraints that can be used to better construct complex in-depth questions and verify answers: (1) functional dependencies can be used to pinpoint critical keywords that an LLM must know to properly answer a given question containing certain attribute values; and (2) foreign key constraints can be used to join relations and construct multi-hop questions, which can be arbitrarily long and used to debug intermediate answers. We thus propose ERBench, which uses these integrity constraints to convert any database into an LLM benchmark. ERBench supports continuous evaluation as databases change, multimodal questions, and various prompt engineering techniques. In our experiments, we construct LLM benchmarks using databases of multiple domains and make an extensive comparison of contemporary LLMs. We show how ERBench can properly evaluate any LLM by not only checking for answer correctness, but also effectively verifying the rationales by looking for the right keywords.

The rapid progress in Large Language Models (LLMs) poses potential risks such as generating unethical content. Assessing the values embedded in LLMs' generated responses can help expose their misalignment, but this relies on reference-free value evaluators, e.g. fine-tuned LLMs or closed-source models like GPT-4. Nevertheless, two key challenges emerge in open-ended value evaluation: the evaluator should adapt to changing human value definitions with minimal annotation, against their own bias (adaptability); and remain robust across varying value expressions and scenarios (generalizability). To handle these challenges, we introduce CLAVE, a novel framework that integrates two complementary LLMs: a large model to extract high-level value concepts from diverse responses, leveraging its extensive knowledge and generalizability, and a small model fine-tuned on these concepts to adapt to human value annotations. This dual-model framework enables adaptation to any value system using <100 human-labeled samples per value type. We also present ValEval, a comprehensive dataset comprising 13k+ (text,value,label) tuples across diverse domains, covering three major value systems. We benchmark the performance of 15+ popular LLM evaluators and fully analyze their strengths and weaknesses. Our findings reveal that CLAVE combining a large prompt-based model and a small fine-tuned one serves as an optimal balance in value evaluation.

The distribution shifts between training and test data typically undermine the performance of models. In recent years, lots of work pays attention to domain generalization (DG) where distribution shifts exist and target data are unseen. Despite the progress in algorithm design, two foundational factors have long been ignored: 1) the optimization for regularization-based objectives, and 2) the model selection for DG since no knowledge about the target domain can be utilized. In this paper, we propose Mixup guided optimization and selection techniques for DG. For optimization, we utilize an adapted Mixup to generate an out-of-distribution dataset that can guide the preference direction and optimize with Pareto optimization. For model selection, we generate a validation dataset with a closer distance to the target distribution, and thereby it can better represent the target data. We also present some theoretical insights behind our proposals. Comprehensive experiments demonstrate that our model optimization and selection techniques can largely improve the performance of existing domain generalization algorithms and even achieve new state-of-the-art results.

Recommendation systems guide users in locating their desired information within extensive content repositories. Usually, a recommendation model is optimized to enhance accuracy metrics from a user utility standpoint, such as click-through rate or matching relevance. However, a responsible industrial recommendation model must address not only user utility (responsibility to users) but also other objectives, including increasing platform revenue (responsibility to platforms), ensuring fairness (responsibility to content creators), and maintaining unbiasedness (responsibility to long-term healthy development). Multi-objective learning is a promising approach for achieving responsible recommendation models. Nevertheless, current methods encounter two challenges: difficulty in scaling to heterogeneous objectives within a unified framework, and inadequate controllability over objective priority during optimization, leading to uncontrollable solutions.

We investigate node representation learning on text-attributed graphs (TAGs), where nodes are associated with text information. Although recent studies on graph neural networks (GNNs) and pretrained language models (PLMs) have exhibited their power in encoding network and text signals, respectively, less attention has been paid to delicately coupling these two types of models on TAGs. Specifically, existing GNNs rarely model text in each node in a contextualized way; existing PLMs can hardly be applied to characterize graph structures due to their sequence architecture. To address these challenges, we propose HASH-CODE, a High-frequency Aware Spectral Hierarchical Contrastive Selective Coding method that integrates GNNs and PLMs into a unified model. Different from previous "cascaded architectures" that directly add GNN layers upon a PLM, our HASH-CODE relies on five self-supervised optimization objectives to facilitate thorough mutual enhancement between network and text signals in diverse granularities. Moreover, we show that existing contrastive objective learns the low-frequency component of the augmentation graph and propose a high-frequency component (HFC)-aware contrastive learning objective that makes the learned embeddings more distinctive. Extensive experiments on six real-world benchmarks substantiate the efficacy of our proposed approach. In addition, theoretical analysis and item embedding visualization provide insights into our model interoperability.

The evaluation of large language models (LLMs) is crucial to assess their performance and mitigate potential security risks. In this paper, we introduce PromptBench, a unified library to evaluate LLMs. It consists of several key components that can be easily used and extended by researchers: prompt construction, prompt engineering, dataset and model loading, adversarial prompt attack, dynamic evaluation protocols, and analysis tools. PromptBench is designed as an open, general, and flexible codebase for research purpose. It aims to facilitate original study in creating new benchmarks, deploying downstream applications, and designing new evaluation protocols. The code is available at: https://github.com/microsoft/promptbench and will be continuously supported.

Time series remains one of the most challenging modalities in machine learning research. Out-of-distribution (OOD) detection and generalization on time series often face difficulties due to their non-stationary nature, wherein the distribution changes over time. Thedynamicdistributions within time series present significant challenges for existing algorithms, especially in identifying invariant distributions, as most focus on scenarios where domain information is provided as prior knowledge. This paper aims to address the issues induced by non-stationarity in time series through the exploration of subdomains within a complete dataset for generalized representation learning. We proposeDiversify, a general framework, for OOD detection and generalization on dynamic distributions of time series.Diversifyoperates through an iterative process: first identifying the’worst-case’latent distribution scenario, then working to minimize the gaps between these latent distributions. We implementDiversifyby combining existing OOD detection methods according to either extracted features or outputs of models for detection while we also directly utilize outputs for classification. Theoretical insights support the framework's validity. Extensive experiments are conducted on seven datasets with different OOD settings across gesture recognition, speech commands recognition, wearable stress and affect detection, and sensor-based human activity recognition. Qualitative and quantitative results demonstrate thatDiversifylearns more generalized features and significantly outperforms other baselines.

During the COVID-19 pandemic, forecasting COVID-19 trends to support planning and response was a priority for scientists and decision makers alike. In the United States, COVID-19 forecasting was coordinated by a large group of universities, companies, and government entities led by the Centers for Disease Control and Prevention and the US COVID-19 Forecast Hub (https://covid19forecasthub.org). We evaluated approximately 9.7 million forecasts of weekly state-level COVID-19 cases for predictions 1-4 weeks into the future submitted by 24 teams from August 2020 to December 2021. We assessed coverage of central prediction intervals and weighted interval scores (WIS), adjusting for missing forecasts relative to a baseline forecast, and used a Gaussian generalized estimating equation (GEE) model to evaluate differences in skill across epidemic phases that were defined by the effective reproduction number. Overall, we found high variation in skill across individual models, with ensemble-based forecasts outperforming other approaches. Forecast skill relative to the baseline was generally higher for larger jurisdictions (e.g., states compared to counties). Over time, forecasts generally performed worst in periods of rapid changes in reported cases (either in increasing or decreasing epidemic phases) with 95% prediction interval coverage dropping below 50% during the growth phases of the winter 2020, Delta, and Omicron waves. Ideally, case forecasts could serve as a leading indicator of changes in transmission dynamics. However, while most COVID-19 case forecasts outperformed a naïve baseline model, even the most accurate case forecasts were unreliable in key phases. Further research could improve forecasts of leading indicators, like COVID-19 cases, by leveraging additional real-time data, addressing performance across phases, improving the characterization of forecast confidence, and ensuring that forecasts were coherent across spatial scales. In the meantime, it is critical for forecast users to appreciate current limitations and use a broad set of indicators to inform pandemic-related decision making.

Exposed surface for buildings (ESB), which refers to exposed surfaces with traces of building construction, often leads to urban dust. Accurate ESB detection is important for planning urban development and improving urban environment. Fine-grained monitoring of ESB typically needs massive high-quality pixel-level labels, which are demanding and expensive. In contrast, obtaining cost-efficient image-level labels is more promising. Most image-level weakly supervised methods can extract pixel-level pseudo labels using the class activation map (CAM) generated by the classification network. Subsequently, these labels are applied to train the semantic segmentation network. However, the CAM is easy to miss fine-grained information, which leads to label noise. Moreover, the downsampling in the segmentation networks will further loss the spatial information. Furthermore, the sparse distribution and irregular shape of ESB pose additional challenges. Given these problems, we propose a stepwise refining image-level weakly supervised semantic segmentation method (SRIWS): 1) we introduce a new data augmentation method called SRMix to oversample the classification dataset; 2) we propose a two-branch network with a superpixel pooling layer (SPNet) as the semantic segmentation network to capture both global semantic information and spatial details; and 3) to alleviate the impact of potential noise in the initial labels, we design the high-confidence sample filtering operation (HSF) during the SPNet training. The evaluation experiments for the SRIWS were performed on three datasets. The results confirm that our proposed SRIWS presents a superior performance in recognizing ESB compared with existing state-of-the-art methods. In addition, numerous ablation experimental results indicate the effectiveness and robustness of our SRIWS.

Most of the existing deep learning-based hyperspectral image (HSI) classification algorithms are based on supervised learning, where large number of annotated labels with high acquisition cost are required. Self-supervised learning (SSL) methods can learn abundant representations using large amount of unlabeled data, thereby reducing the reliability of labels. Particularly, SSL based on Masked Image Modeling (MIM) can extract fine-grained features, which is well-suited for HSI classification as a pixel-level interpretation task. However, MIM has scarcely been investigated in HSI classification. Current algorithms lack a comprehensive consideration of the multiscale spectral-spatial characteristics of HSI when constructing the pre-training task, and there exists high computational cost and redundancy when applied to large-scale HSI. Therefore, this paper develops an SSL framework based on Spectral-Spatial Hierarchical Masked Modeling (S2HM2) for large-scale HSI classification. Considering the spectral-spatial characteristics of HSI, 3D masking strategy and spectral-spatial consistency loss are proposed to construct MIM task. To fully exploit features at each scale, hierarchical 3D Feature Pyramid Network (3D-FPN) is designed as decoder for both pre-text and downstream tasks in a “pixel-to-pixel” manner. In addition, Multi-Scale Masked Feature Modeling (MS-MFM) task is proposed to further facilitate the multiscale feature learning. The SSL pre-training is guided by both MIM and MS-MFM. Experimental results on two large-scale hyperspectral datasets, i.e., WHU-OHS and WHU-H2SR, demonstrate the superiority of the proposed method. Furthermore, transfer learning experiments are conducted on a variety of hyperspectral datasets, where classification accuracies are boosted in most of the scenarios. Source code will be made available at https://github.com/tulilin/S2HM2.

Large language models (LLMs) are gaining increasing popularity in both academia and industry, owing to their unprecedented performance in various applications. As LLMs continue to play a vital role in both research and daily use, their evaluation becomes increasingly critical, not only at the task level, but also at the society level for better understanding of their potential risks. Over the past years, significant efforts have been made to examine LLMs from various perspectives. This paper presents a comprehensive review of these evaluation methods for LLMs, focusing on three key dimensions: what to evaluate , where to evaluate , and how to evaluate . Firstly, we provide an overview from the perspective of evaluation tasks, encompassing general natural language processing tasks, reasoning, medical usage, ethics, education, natural and social sciences, agent applications, and other areas. Secondly, we answer the ‘where’ and ‘how’ questions by diving into the evaluation methods and benchmarks, which serve as crucial components in assessing the performance of LLMs. Then, we summarize the success and failure cases of LLMs in different tasks. Finally, we shed light on several future challenges that lie ahead in LLMs evaluation. Our aim is to offer invaluable insights to researchers in the realm of LLMs evaluation, thereby aiding the development of more proficient LLMs. Our key point is that evaluation should be treated as an essential discipline to better assist the development of LLMs. We consistently maintain the related open-source materials at: https://github.com/MLGroupJLU/LLM-eval-survey

In this article, we study the problem of personalized news headline generation, which aims to produce not only concise and fact-consistent titles for news articles but also decorate these titles as personalized irresistible reading invitations by incorporating readers’ preferences. We propose an approach named PNG ( P ersonalized N ews headline G enerator) by utilizing distant supervision in readers’ past click behaviors to resolve. First, user preference representations are learned through a knowledge-aware user encoder that comprehensively captures the genuine, sequential, and flash interests of users reflected in their historical clicked news. Then, a user-perturbed pointer-generator network is devised to accomplish the headline generation in which the learned user representations implicitly affect the word prediction. The proposed model is optimized by reinforcement learning solvers where indicators on factual, personalized, and linguistic aspects of the generated headline are regarded as rewards. Extensive experiments are conducted on the real-world dataset PENS, 1 which is a large-scale benchmark collected from Microsoft News. Both the quantitative and qualitative results validate the effectiveness of our approach.

Federated learning (FL) has attracted increasing attention to building models without accessing raw user data, especially in healthcare. In real applications, different federations can seldom work together due to possible reasons such as data heterogeneity and distrust/inexistence of the central server. In this article, we propose a novel framework called MetaFed to facilitate trustworthy FL between different federations. MetaFed obtains a personalized model for each federation without a central server via the proposed cyclic knowledge distillation. Specifically, MetaFed treats each federation as a meta distribution and aggregates knowledge of each federation in a cyclic manner. The training is split into two parts: common knowledge accumulation and personalization. Comprehensive experiments on seven benchmarks demonstrate that MetaFed without a server achieves better accuracy compared with state-of-the-art methods [e.g., 10%+ accuracy improvement compared with the baseline for physical activity monitoring dataset (PAMAP2)] with fewer communication costs. More importantly, MetaFed shows remarkable performance in real-healthcare-related applications.

Social network alignment, aiming at linking identical identities across different social platforms, is a fundamental task in social graph mining. Most existing approaches are supervised models and require a large number of manually labeled data, which are infeasible in practice considering the yawning gap between social platforms. Recently, isomorphism across social networks is incorporated as complementary to link identities from the distribution level, which contributes to alleviating the dependency on sample-level annotations. Adversarial learning is adopted to learn a shared projection function by minimizing the distance between two social distributions. However, the hypothesis of isomorphism might not always hold true as social user behaviors are generally unpredictable, and thus a shared projection function is insufficient to handle the sophisticated cross-platform correlations. In addition, adversarial learning suffers from training instability and uncertainty, which may hinder model performance. In this article, we propose a novel meta-learning-based social network alignment model Meta-SNA to effectively capture the isomorphism and the unique characteristics of each identity. Our motivation lies in learning a shared meta-model to preserve the global cross-platform knowledge and an adaptor to learn a specific projection function for each identity. Sinkhorn distance is further introduced as the distribution closeness measurement to tackle the limitations of adversarial learning, which owns an explicitly optimal solution and can be efficiently computed by the matrix scaling algorithm. Empirically, we evaluate the proposed model over multiple datasets, and the experimental results demonstrate the superiority of Meta-SNA.

In this paper, we move towards combining large parametric models with non-parametric prototypical networks. We propose prototypical fine-tuning, a novel prototypical framework for fine-tuning pretrained language models (LM), which automatically learns a bias to improve predictive performance for varying data sizes, especially low-resource settings. Our prototypical fine-tuning approach can automatically adjust the model capacity according to the number of data points and the model's inherent attributes. Moreover, we propose four principles for effective prototype fine-tuning towards the optimal solution. Experimental results across various datasets show that our work achieves significant performance improvements under various low-resource settings, as well as comparable and usually better performances in high-resource scenarios.

Rui Li, Xu Chen, Chaozhuo Li, Yanming Shen, Jianan Zhao, Yujing Wang, Weihao Han, Hao Sun, Weiwei Deng, Qi Zhang, Xing Xie. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Self-training (ST) has prospered again in language understanding by augmenting the finetuning of big pre-trained models when labeled data is insufficient.However, it remains challenging to incorporate ST into attributecontrollable language generation.Augmented only by self-generated pseudo text, generation models over-exploit the previously learned text space and fail to explore a larger one, suffering from a restricted generalization boundary and limited controllability.In this work, we propose DuNST, a novel ST framework to tackle these problems.DuNST jointly models text generation and classification as a dual process and further perturbs and escapes from the collapsed space by adding two kinds of flexible noise.In this way, our model could construct and utilize both pseudo text generated from given labels and pseudo labels predicted from available unlabeled text, which are gradually refined during the ST phase.Theoretically, we show that DuNST can be viewed as enhancing the exploration of the potentially larger real text space while maintaining exploitation, guaranteeing improved performance.Experiments on three controllable generation tasks show that DuNST significantly boosts control accuracy with comparable generation fluency and diversity against several strong baselines.

Wenjun Peng, Jingwei Yi, Fangzhao Wu, Shangxi Wu, Bin Bin Zhu, Lingjuan Lyu, Binxing Jiao, Tong Xu, Guangzhong Sun, Xing Xie. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Sequential recommendation (SR) aims to model users' dynamic preferences from a series of interactions. A pivotal challenge in user modeling for SR lies in the inherent variability of user preferences. An effective SR model is expected to capture both the long-term and short-term preferences exhibited by users, wherein the former can offer a comprehensive understanding of stable interests that impact the latter. To more effectively capture such information, we incorporate locality inductive bias into the Transformer by amalgamating its global attention mechanism with a local convolutional filter, and adaptively ascertain the mixing importance on a personalized basis through layer-aware adaptive mixture units, termed as AdaMCT. Moreover, as users may repeatedly browse potential purchases, it is expected to consider multiple relevant items concurrently in long-/short-term preferences modeling. Given that softmax-based attention may promote unimodal activation, we propose the Squeeze-Excitation Attention (with sigmoid activation) into SR models to capture multiple pertinent items (keys) simultaneously. Extensive experiments on three widely employed benchmarks substantiate the effectiveness and efficiency of our proposed approach. Source code is available at https://github.com/juyongjiang/AdaMCT.

Federated learning (FL) is important for privacy-preserving services by training models without collecting raw user data. Most FL algorithms assume all data is annotated, which is impractical due to the high cost of labeling data in real applications. To alleviate the reliance on labeled data, semi-supervised federated learning (SSFL) has been proposed to utilize unlabeled data on clients to improve model performance. However, most existing methods either have privacy issues which share models trained on other clients, or generate pseudo-labels for unlabeled local datasets with the global model, which is usually biased towards the global data distribution. The latter may lead to sub-optimal accuracy of pseudo-labels, due to the gap between the local data distribution and the global model, especially in non-IID settings. In this paper, we propose a semi-supervised heterogeneous federated learning method with local knowledge enhancement, called FedLoKe, which aims to train an accurate global model from both labeled and unlabeled local data with non-IID distributions. Specifically, in FedLoKe, the server maintains a global model to capture global data distribution, and each client learns a local model to capture local data distribution. Since the distribution captured by the local model is aligned with the local data distribution, we utilize it to generate high-accuracy pseudo-labels of the unlabeled dataset for global model training. To prevent the local model from severely overfitting the small number of local labeled data, we further use the exponential moving average and apply the global model to generate pseudo-labels for local modeling training. Experiments on four datasets show the effectiveness of FedLoKe. Our code is available at: https://github.com/zcfinal/FedLoKe.

Warning: this paper includes model outputs showing offensive content.Recent large-scale Visual-Language Generative Models (VLGMs) have achieved unprecedented improvement in multimodal image/text generation.However, these models might also generate toxic content, e.g., offensive text and pornography images, raising significant ethical risks.Despite exhaustive studies on toxic degeneration of language models, this problem remains largely unexplored within the context of visual-language generation.This work delves into the propensity for toxicity generation and susceptibility to toxic data across various VLGMs.For this purpose, we built ToViLaG, a dataset comprising 32K co-toxic/mono-toxic text-image pairs and 1K innocuous but evocative text that tends to stimulate toxicity.Furthermore, we propose WInToRe, a novel toxicity metric tailored to visual-language generation, which theoretically reflects different aspects of toxicity considering both input and output.On such a basis, we benchmarked the toxicity of a diverse spectrum of VLGMs and discovered that some models do more evil than expected while some are more vulnerable to infection, underscoring the necessity of VLGMs detoxification.Therefore, we develop an innovative information bottleneckbased detoxification method.Our method reduces toxicity while maintaining acceptable generation quality, providing a promising initial solution to this line of research.* Work done as an intern at MSRA mentored by X. Yi.

Pre-trained language models (PLMs) have achieved the preeminent position in dense retrieval due to their powerful capacity in modeling intrinsic semantics.However, most existing PLM-based retrieval models encounter substantial computational costs and are infeasible for processing long documents.In this paper, a novel retrieval model Longtriever is proposed to embrace three core challenges of long document retrieval: substantial computational cost, incomprehensive document understanding, and scarce annotations.Longtriever splits long documents into short blocks and then efficiently models the local semantics within a block and the global context semantics across blocks in a tightly-coupled manner.A pretraining phase is further proposed to empower Longtriever to achieve a better understanding of underlying semantic correlations.Experimental results on two popular benchmark datasets demonstrate the superiority of our proposal.The source code is released at https: //github.com/SamuelYang1/Longtriever .

Federated learning is used to train a shared model in a decentralized way without clients sharing private data with each other. Federated learning systems are susceptible to poisoning attacks when malicious clients send false updates to the central server. Existing defense strategies are ineffective under non-IID data settings. This paper proposes a new defense strategy, FedCPA (Federated learning with Critical Parameter Analysis). Our attack-tolerant aggregation method is based on the observation that benign local models have similar sets of top-k and bottom-k critical parameters, whereas poisoned local models do not. Experiments with different attack scenarios on multiple datasets demonstrate that our model outperforms existing defense strategies in defending against poisoning attacks.

Deep neural networks are susceptible to adversarial examples, posing a significant security risk in critical applications. Adversarial Training (AT) is a well-established technique to enhance adversarial robustness, but it often comes at the cost of decreased generalization ability. This paper proposes Robustness Critical Fine-Tuning (RiFT), a novel approach to enhance generalization without compromising adversarial robustness. The core idea of RiFT is to exploit the redundant capacity for robustness by fine-tuning the adversarially trained model on its non-robust-critical module. To do so, we introduce module robust criticality (MRC), a measure that evaluates the significance of a given module to model robustness under worst-case weight perturbations. Using this measure, we identify the module with the lowest MRC value as the non-robust-critical module and fine-tune its weights to obtain fine-tuned weights. Subsequently, we linearly interpolate between the adversarially trained weights and fine-tuned weights to derive the optimal fine-tuned model weights. We demonstrate the efficacy of RiFT on ResNet18, ResNet34, and WideResNet34-10 models trained on CIFAR10, CIFAR100, and Tiny-ImageNet datasets. Our experiments show that RiFT can significantly improve both generalization and out-of-distribution robustness by around 1.5% while maintaining or even slightly enhancing adversarial robustness. Code is available at https://github.com/Immortalise/RiFT.

To solve the information explosion problem and enhance user experience in various online applications, recommender systems have been developed to model users’ preferences. Although numerous efforts have been made toward more personalized recommendations, recommender systems still suffer from several challenges, such as data sparsity and cold-start problems. In recent years, generating recommendations with the knowledge graph as side information has attracted considerable interest. Such an approach can not only alleviate the above mentioned issues for a more accurate recommendation, but also provide explanations for recommended items. In this paper, we conduct a systematical survey of knowledge graph-based recommender systems. We collect recently published papers in this field, and group them into three categories, i.e., embedding-based methods, connection-based methods, and propagation-based methods. Also, we further subdivide each category according to the characteristics of these approaches. Moreover, we investigate the proposed algorithms by focusing on how the papers utilize the knowledge graph for accurate and explainable recommendation. Finally, we propose several potential research directions in this field.

Self-training (ST) has come to fruition in language understanding tasks by producing pseudo labels, which reduces the labeling bottleneck of language model fine-tuning. Nevertheless, in facilitating semi-supervised controllable language generation, ST faces two key challenges. First, augmented by self-generated pseudo text, generation models tend to over-exploit the previously learned text distribution, suffering from mode collapse and poor generation diversity. Second, generating pseudo text in each iteration is time-consuming, severely decelerating the training process. In this work, we propose KEST, a novel and efficient self-training framework to handle these problems. KEST utilizes a kernel-based loss, rather than standard cross entropy, to learn from the soft pseudo text produced by a shared non-autoregressive generator. We demonstrate both theoretically and empirically that KEST can benefit from more diverse pseudo text in an efficient manner, which allows not only refining and exploiting the previously fitted distribution but also enhanced exploration towards a larger potential text space, providing a guarantee of improved performance. Experiments on three controllable generation tasks demonstrate that KEST significantly improves control accuracy while maintaining comparable text fluency and generation diversity against several strong baselines.

Federated learning (FL) is an important technique for learning models from decentralized data in a privacy-preserving way. Existing FL methods usually uniformly sample clients for local model learning in each round. However, different clients may have significantly different data sizes, and the clients with more data cannot have more opportunities to contribute to model training, which may lead to inferior performance. In this paper, instead of client uniform sampling, we propose a novel data uniform sampling strategy for federated learning (FedSampling), which can effectively improve the performance of federated learning especially when client data size distribution is highly imbalanced across clients. In each federated learning round, local data on each client is randomly sampled for local model learning according to a probability based on the server desired sample size and the total sample size on all available clients. Since the data size on each client is privacy-sensitive, we propose a privacy-preserving way to estimate the total sample size with a differential privacy guarantee. Experiments on four benchmark datasets show that FedSampling can effectively improve the performance of federated learning.

Federated learning enables learning from decentralized data sources without compromising privacy, which makes it a crucial technique. However, it is vulnerable to model poisoning attacks, where malicious clients interfere with the training process. Previous defense mechanisms have focused on the server-side by using careful model aggregation, but this may not be effective when the data is not identically distributed or when attackers can access the information of benign clients. In this paper, we propose a new defense mechanism that focuses on the client-side, called FedDefender, to help benign clients train robust local models and avoid the adverse impact of malicious model updates from attackers, even when a server-side defense cannot identify or remove adversaries. Our method consists of two main components: (1) attack-tolerant local meta update and (2) attack-tolerant global knowledge distillation. These components are used to find noise-resilient model parameters while accurately extracting knowledge from a potentially corrupted global model. Our client-side defense strategy has a flexible structure and can work in conjunction with any existing server-side strategies. Evaluations of real-world scenarios across multiple datasets show that the proposed method enhances the robustness of federated learning against model poisoning attacks.

The ease of spreading false information online enables individuals with malicious intent to manipulate public opinion and destabilize social stability. Recently, fake news detection based on evidence retrieval has gained popularity in an effort to identify fake news reliably and reduce its impact. Evidence retrieval-based methods can improve the reliability of fake news detection by computing the textual consistency between the evidence and the claim in the news. In this paper, we propose a framework for fake news detection based on MUlti- Step Evidence Retrieval enhancement (MUSER), which simulates the steps of human beings in the process of reading news, summarizing, consulting materials, and inferring whether the news is true or fake. Our model can explicitly model dependencies among multiple pieces of evidence, and perform multi-step associations for the evidence required for news verification through multi-step retrieval. In addition, our model is able to automatically collect existing evidence through paragraph retrieval and key evidence selection, which can save the tedious process of manual evidence collection. We conducted extensive experiments on real-world datasets in different languages, and the results demonstrate that our proposed model outperforms state-of-the-art baseline methods for detecting fake news by at least 3% in F1-Macro and 4% in F1-Micro. Furthermore, it provides interpretable evidence for end users.

Human activity recognition (HAR) is a time series classification task that focuses on identifying the motion patterns from human sensor readings. Adequate data is essential but a major bottleneck for training a generalizable HAR model, which assists customization and optimization of online web applications. However, it is costly in time and economy to collect large-scale labeled data in reality, i.e., the low-resource challenge. Meanwhile, data collected from different persons have distribution shifts due to different living habits, body shapes, age groups, etc. The low-resource and distribution shift challenges are detrimental to HAR when applying the trained model to new unseen subjects. In this paper, we propose a novel approach called Diverse and Discriminative representation Learning (DDLearn) for generalizable low-resource HAR. DDLearn simultaneously considers diversity and discrimination learning. With the constructed self-supervised learning task, DDLearn enlarges the data diversity and explores the latent activity properties. Then, we propose a diversity preservation module to preserve the diversity of learned features by enlarging the distribution divergence between the original and augmented domains. Meanwhile, DDLearn also enhances semantic discrimination by learning discriminative representations with supervised contrastive learning. Extensive experiments on three public HAR datasets demonstrate that our method significantly outperforms state-of-art methods by an average accuracy improvement of 9.5% under the low-resource distribution shift scenarios, while being a generic, explainable, and flexible framework. Code is available at: https://github.com/microsoft/robustlearn.

The nucleus of online sponsored search systems lies in measuring the relevance between the search intents of users and the advertising purposes of advertisers. Existing conventional doublet-based (query-keyword) relevance models solely rely on short queries and keywords to uncover such intents, which ignore the diverse and personalized preferences of participants (i.e., users and advertisers), resulting in undesirable advertising performance. In this paper, we investigate the novel problem of Personalized A dvertiser-aware Sponsored Search (PASS). Our motivation lies in incorporating the portraits of users and advertisers into relevance models to facilitate the modeling of intrinsic search intents and advertising purposes, leading to a quadruple-based (i.e., user-query-keyword-advertiser) task. Various types of historical behaviors are explored in the format of hypergraphs to provide abundant signals on identifying the preferences of participants. A novel heterogeneous textual hypergraph transformer is further proposed to deeply fuse the textual semantics and the high-order hypergraph topology. Our proposal is extensively evaluated over real industry datasets, and experimental results demonstrate its superiority.

Machine learning is becoming increasingly important in today's world. Beyond its powerful performances, there has been an emerging concern about the trustworthiness of machine learning, including but not limited to: robustness to malicious attacks, generalization to unseen datasets, and interpretability to explain its outputs. Such concerns are even more urgent in some safety-critical applications such as medical diagnosis and autonomous driving. Trustworthy machine learning (TrustML) aims to tackle these challenges from the perspectives of theory, algorithm, and applications. In this tutorial, we will give a comprehensive introduction to the recent advance of trustworthy machine learning in robustness, generalization, and interpretability. We will cover their problem formulation, related research, popular algorithms, and successful applications. Additionally, we will also introduce some potential challenges for future research. We do hope that this tutorial will not only serve as a platform to understand TrustML, but also raise the awareness of everyone for more trustworthy applications.

This paper introduces a unified causal lens for understanding representative model interpretation methods. We show that their explanation scores align with the concept of average treatment effect in causal inference, which allows us to evaluate their relative strengths and limitations from a unified causal perspective. Based on our observations, we outline the major challenges in applying causal inference to model interpretation, including identifying common causes that can be generalized across instances and ensuring that explanations provide a complete causal explanation of model predictions. We then present CIMI, a Causality-Inspired Model Interpreter, which addresses these challenges. Our experiments show that CIMI provides more faithful and generalizable explanations with improved sampling efficiency, making it particularly suitable for larger pretrained models.

News recommendation is essential for personalized news distribution. Federated news recommendation, which enables collaborative model learning from multiple clients without sharing their raw data, is a promising approach for preserving users' privacy. However, the security of federated news recommendation is still unclear. In this paper, we study this problem by proposing an untargeted attack on federated news recommendation called UA-FedRec. By exploiting the prior knowledge of news recommendation and federated learning, UA-FedRec can effectively degrade the model performance with a small percentage of malicious clients. First, the effectiveness of news recommendation highly depends on user modeling and news modeling. We design a news similarity perturbation method to make representations of similar news farther and those of dissimilar news closer to interrupt news modeling, and propose a user model perturbation method to make malicious user updates in opposite directions of benign updates to interrupt user modeling. Second, updates from different clients are typically aggregated with a weighted average based on their sample sizes. We propose a quantity perturbation method to enlarge sample sizes of malicious clients in a reasonable range to amplify the impact of malicious updates. Extensive experiments on two real-world datasets show that UA-FedRec can effectively degrade the accuracy of existing federated news recommendation methods, even when defense is applied. Our study reveals a critical security issue in existing federated news recommendation systems and calls for research efforts to address the issue. Our code is available at https://github.com/yjw1029/UA-FedRec.

Domain generalization (DG) approaches typically use the hypothesis learned on source domains for inference on the unseen target domain. However, such a hypothesis can be arbitrarily far from the optimal one for the target domain, induced by a gap termed ''adaptivity gap.'' Without exploiting the domain information from the unseen test samples, adaptivity gap estimation and minimization are intractable, which hinders us to robustify a model to any unknown distribution. In this paper, we first establish a generalization bound that explicitly considers the adaptivity gap. Our bound motivates two strategies to reduce the gap: the first one is ensembling multiple classifiers to enrich the hypothesis space, then we propose effective gap estimation methods for guiding the selection of a better hypothesis for the target. The other method is minimizing the gap directly by adapting model parameters using online target samples. We thus propose Domain-specific Risk Minimization (DRM). During training, DRM models the distributions of different source domains separately; for inference, DRM performs online model steering using the source hypothesis for each arriving target sample. Extensive experiments demonstrate the effectiveness of the proposed DRM for domain generalization. Code is available at: https://github.com/yfzhang114/AdaNPC.

Recently, the bias-related issues in GNN-based link prediction have raised widely spread concerns. In this paper, we emphasize the bias on links across different node clusters, which we call cross-links, after considering its significance in both easing information cocoons and preserving graph connectivity. Instead of following the objective-oriented mechanism in prior works with compromised utility, we empirically find that existing GNN models face severe data bias between internal-links (links within the same cluster) and cross-links, and this inspires us to rethink the bias issue on cross-links from a data perspective. Specifically, we design a simple yet effective twin-structure framework, which can be easily applied to most of GNNs to mitigate the bias as well as boost their utility in an end-to-end manner. The basic idea is to generate debiased node embeddings as demonstrations, and fuse them into the embeddings of original GNNs. In particular, we learn debiased node embeddings with the help of augmented supervision signals, and a novel dynamic training strategy is designed to effectively fuse debiased node embeddings with the original node embeddings. Experiments on three datasets with six common GNNs show that our framework can not only alleviate the bias between internal-links and cross-links, but also boost the overall accuracy. Comparisons with other state-of-the-art methods also verify the superiority of our method.

GNN explanation method aims to identify an explanatory subgraph which contains the most informative components of the full graph. However, a major limitation of existing GNN explainers is that they are not robust to the structurally corrupted graphs, e.g., graphs with noisy or adversarial edges. On the one hand, existing GNN explainers mostly explore explanations based on either the raw graph features or the learned latent representations, both of which can be easily corrupted. On the other hand, the corruptions in graphs are irregular in terms of the structural properties, e.g., the size or connectivity of graphs, which makes the rigorous constraints used by previous GNN explainers unfeasible. To address these issues, we propose a robust GNN explainer called V-InfoR. Specifically, a robust graph representation extractor, which takes insights of variational inference, is proposed to infer the latent distribution of graph representations. Instead of directly using the corrupted raw features or representations of each single graph, we sample the graph representations from the inferred distribution for the downstream explanation generator, which can effectively eliminate the minor corruption. We next formulate the explanation exploration as a graph information bottleneck (GIB) optimization problem. As a more general method that does not need any rigorous structural constraints, our GIB-based method can adaptively capture both the regularity and irregularity of the severely corrupted graphs for explanation. Extensive evaluations on both synthetic and real-world datasets indicate that V-InfoR significantly improves the GNN explanation performance for the structurally corrupted graphs. Code and dataset are available at https://anonymous.4open.science/r/V-InfoR-EF88

Text-attributed graphs (TAGs) are prevalent in various real-world scenarios, where each node is associated with a text description. The cornerstone of representation learning on TAGs lies in the seamless integration of textual semantics within individual nodes and the topological connections across nodes. Recent advancements in pre-trained language models (PLMs) and graph neural networks (GNNs) have facilitated effective learning on TAGs, garnering increased research interest. However, the absence of meaningful benchmark datasets and standardized evaluation procedures for TAGs has impeded progress in this field. In this paper, we propose CS-TAG, a comprehensive and diverse collection of challenging benchmark datasets for TAGs. The CS-TAG datasets are notably large in scale and encompass a wide range of domains, spanning from citation networks to purchase graphs. In addition to building the datasets, we conduct extensive benchmark experiments over CS-TAG with various learning paradigms, including PLMs, GNNs, PLM-GNN co-training methods, and the proposed novel topological pre-training of language models. In a nutshell, we provide an overview of the CS-TAG datasets, standardized evaluation procedures, and present baseline experiments. The entire CS-TAG project is publicly accessible at \url{https://github.com/sktsherlock/TAG-Benchmark}.

This paper studies the problem of active causal discovery when the experiments can be done based on multi-fidelity oracles, where higher fidelity experiments are more precise and expensive, while the lower ones are cheaper but less accurate. In this paper, we formally define the task of multi-fidelity active causal discovery, and design a probabilistic model for solving this problem. In specific, we first introduce a mutual-information based acquisition function to determine which variable should be intervened at which fidelity, and then a cascading model is proposed to capture the correlations between different fidelity oracles. Beyond the above basic framework, we also extend it to the batch intervention scenario. We find that the theoretical foundations behind the widely used and efficient greedy method do not hold in our problem. To solve this problem, we introduce a new concept called $\epsilon$-submodular, and design a constraint based fidelity model to theoretically validate the greedy method. We conduct extensive experiments to demonstrate the effectiveness of our model.

Embedding-based retrieval methods construct vector indices to search for document representations that are most similar to the query representations. They are widely used in document retrieval due to low latency and decent recall performance. Recent research indicates that deep retrieval solutions offer better model quality, but are hindered by unacceptable serving latency and the inability to support document updates. In this paper, we aim to enhance the vector index with end-to-end deep generative models, leveraging the differentiable advantages of deep retrieval models while maintaining desirable serving efficiency. We propose Model-enhanced Vector Index (MEVI), a differentiable model-enhanced index empowered by a twin-tower representation model. MEVI leverages a Residual Quantization (RQ) codebook to bridge the sequence-to-sequence deep retrieval and embedding-based models. To substantially reduce the inference time, instead of decoding the unique document ids in long sequential steps, we first generate some semantic virtual cluster ids of candidate documents in a small number of steps, and then leverage the well-adapted embedding vectors to further perform a fine-grained search for the relevant documents in the candidate virtual clusters. We empirically show that our model achieves better performance on the commonly used academic benchmarks MSMARCO Passage and Natural Questions, with comparable serving latency to dense retrieval solutions.

Relevance models measure the semantic closeness between queries and the candidate ads, widely recognized as the nucleus of sponsored search systems. Conventional relevance models solely rely on the textual data within the queries and ads, whose performance is hindered by the scarce semantic information in these short texts. Recently, user behavior graphs have been incorporated to provide complementary information beyond pure textual semantics.Despite the promising performance, behavior-enhanced models suffer from exhausting resource costs due to the extra computations introduced by explicit topological aggregations. In this paper, we propose a novel Multi-Grained Topological Pre-Training paradigm, MGTLM, to teach language models to understand multi-grained topological information in behavior graphs, which contributes to eliminating explicit graph aggregations and avoiding information loss. Extensive experimental results over online and offline settings demonstrate the superiority of our proposal.

Many real-world graph learning tasks require handling dynamic graphs where new nodes and edges emerge. Dynamic graph learning methods commonly suffer from the catastrophic forgetting problem, where knowledge learned for previous graphs is overwritten by updates for new graphs. To alleviate the problem, continual graph learning methods are proposed. However, existing continual graph learning methods aim to learn new patterns and maintain old ones with the same set of parameters of fixed size, and thus face a fundamental tradeoff between both goals. In this paper, we propose Parameter Isolation GNN (PI-GNN) for continual learning on dynamic graphs that circumvents the tradeoff via parameter isolation and expansion. Our motivation lies in that different parameters contribute to learning different graph patterns. Based on the idea, we expand model parameters to continually learn emerging graph patterns. Meanwhile, to effectively preserve knowledge for unaffected patterns, we find parameters that correspond to them via optimization and freeze them to prevent them from being rewritten. Experiments on eight real-world datasets corroborate the effectiveness of PI-GNN compared to state-of-the-art baselines.

Cross-Domain Recommendation (CDR) is capable of incorporating auxiliary information from multiple domains to advance recommendation performance. Conventional CDR methods primarily rely on overlapping users, whereby knowledge is conveyed between the source and target identities belonging to the same natural person. However, such a heuristic assumption is not universally applicable due to an individual may exhibit distinct or even conflicting preferences in different domains, leading to potential noises. In this paper, we view the anchor links between users of various domains as the learnable parameters to learn the task-relevant cross-domain correlations. A novel optimal transport based model ALCDR is further proposed to precisely infer the anchor links and deeply aggregate collaborative signals from the perspectives of intra-domain and inter-domain. Our proposal is extensively evaluated over real-world datasets, and experimental results demonstrate its superiority.

With the availability of massive labeled training data, powerful machine learning models can be trained. However, the traditional I.I.D. assumption that the training and testing data should follow the same distribution is often violated in reality. While existing domain adaptation approaches can tackle domain shift, it relies on the target samples for training. Domain generalization is a promising technology that aims to train models with good generalization ability to unseen distributions. In this tutorial, we will present the recent advance of domain generalization. Specifically, we introduce the background, formulation, and theory behind this topic. Our primary focus is on the methodology, evaluation, and applications. We hope this tutorial can draw interest of the community and provide a thorough review of this area. Eventually, more robust systems can be built for responsible AI. All tutorial materials and updates can be found online at https://dgresearch.github.io/.

Session-based recommendation (SBR) aims to predict the user's next action based on short and dynamic sessions. Recently, there has been an increasing interest in utilizing various elaborately designed graph neural networks (GNNs) to capture the pair-wise relationships among items, seemingly suggesting the design of more complicated models is the panacea for improving the empirical performance. However, these models achieve relatively marginal improvements with exponential growth in model complexity. In this paper, we dissect the classical GNN-based SBR models and empirically find that some sophisticated GNN propagations are redundant, given the readout module plays a significant role in GNN-based models. Based on this observation, we intuitively propose to remove the GNN propagation part, while the readout module will take on more responsibility in the model reasoning process. To this end, we propose the Multi-Level Attention Mixture Network (Atten-Mixer), which leverages both concept-view and instance-view readouts to achieve multi-level reasoning over item transitions. As simply enumerating all possible high-level concepts is infeasible for large real-world recommender systems, we further incorporate SBR-related inductive biases, i.e., local invariance and inherent priority to prune the search space. Experiments on three benchmarks demonstrate the effectiveness and efficiency of our proposal. We also have already launched the proposed techniques to a large-scale e-commercial online service since April 2021, with significant improvements of top-tier business metrics demonstrated in the online experiments on live traffic.

Algorithmic fairness has become an important machine learning problem, especially for mission-critical Web applications. This work presents a self-supervised model, called DualFair, that can debias sensitive attributes like gender and race from learned representations. Unlike existing models that target a single type of fairness, our model jointly optimizes for two fairness criteria—group fairness and counterfactual fairness—and hence makes fairer predictions at both the group and individual levels. Our model uses contrastive loss to generate embeddings that are indistinguishable for each protected group, while forcing the embeddings of counterfactual pairs to be similar. It then uses a self-knowledge distillation method to maintain the quality of representation for the downstream tasks. Extensive analysis over multiple datasets confirms the model’s validity and further shows the synergy of jointly addressing two fairness criteria, suggesting the model’s potential value in fair intelligent Web applications.

Collaborative Filtering (CF) is a widely used and effective technique for recommender systems. In recent decades, there have been significant advancements in latent embedding-based CF methods for improved accuracy, such as matrix factorization, neural collaborative filtering, and LightGCN. However, the explainability of these models has not been fully explored. Adding explainability to recommendation models can not only increase trust in the decision-making process, but also have multiple benefits such as providing persuasive explanations for item recommendations, creating explicit profiles for users and items, and assisting item producers in design improvements.

Deep recommender systems (DRS) are intensively applied in modern web services. To deal with the massive web contents, DRS employs a two-stage workflow: retrieval and ranking, to generate its recommendation results. The retriever aims to select a small set of relevant candidates from the entire items with high efficiency; while the ranker, usually more precise but time-consuming, is supposed to further refine the best items from the retrieved candidates. Traditionally, the two components are trained either independently or within a simple cascading pipeline, which is prone to poor collaboration effect. Though some latest works suggested to train retriever and ranker jointly, there still exist many severe limitations: item distribution shift between training and inference, false negative, and misalignment of ranking order. As such, it remains to explore effective collaborations between retriever and ranker.

Recent recommender systems have shown remarkable performance by using an ensemble of heterogeneous models. However, it is exceedingly costly because it requires resources and inference latency proportional to the number of models, which remains the bottleneck for production. Our work aims to transfer the ensemble knowledge of heterogeneous teachers to a lightweight student model using knowledge distillation (KD), to reduce the huge inference costs while retaining high accuracy. Through an empirical study, we find that the efficacy of distillation severely drops when transferring knowledge from heterogeneous teachers. Nevertheless, we show that an important signal to ease the difficulty can be obtained from the teacher’s training trajectory. This paper proposes a new KD framework, named HetComp, that guides the student model by transferring easy-to-hard sequences of knowledge generated from the teachers’ trajectories. To provide guidance according to the student’s learning state, HetComp uses dynamic knowledge construction to provide progressively difficult ranking knowledge and adaptive knowledge transfer to gradually transfer finer-grained ranking information. Our comprehensive experiments show that HetComp significantly improves the distillation quality and the generalization of the student model.

Graph neural networks (GNNs) have seen widespread usage across multiple real-world applications, yet in transductive learning, they still face challenges in accuracy, efficiency, and scalability, due to the extensive number of trainable parameters in the embedding table and the paradigm of stacking neighborhood aggregations. This paper presents a novel model called xGCN for large-scale network embedding, which is a practical solution for link predictions. xGCN addresses these issues by encoding graph-structure data in an extreme convolutional manner, and has the potential to push the performance of network embedding-based link predictions to a new record. Specifically, instead of assigning each node with a directly learnable embedding vector, xGCN regards node embeddings as static features. It uses a propagation operation to smooth node embeddings and relies on a Refinement neural Network (RefNet) to transform the coarse embeddings derived from the unsupervised propagation into new ones that optimize a training objective. The output of RefNet, which are well-refined embeddings, will replace the original node embeddings. This process is repeated iteratively until the model converges to a satisfying status. Experiments on three social network datasets with link prediction tasks show that xGCN not only achieves the best accuracy compared with a series of competitive baselines but also is highly efficient and scalable.

Deep semantic matching aims at discriminating the relationship between documents based on deep neural networks. In recent years, it becomes increasingly popular to organize documents with a graph structure, then leverage both the intrinsic document features and the extrinsic neighbor features to derive discrimination. Most of the existing works mainly care about how to utilize the presented neighbors, whereas limited effort is made to filter appropriate neighbors. We argue that the neighbor features could be highly noisy and partially useful. Thus, a lack of effective neighbor selection will not only incur a great deal of unnecessary computation cost but also restrict the matching accuracy severely. In this work, we propose a novel framework, C ascaded D eep S emantic M atching ( CDSM ), for accurate and efficient semantic matching on textual graphs. CDSM is highlighted for its two-stage workflow. In the first stage, a lightweight CNN-based ad-hod neighbor selector is deployed to filter useful neighbors for the matching task with a small computation cost. We design both one-step and multi-step selection methods. In the second stage, a high-capacity graph-based matching network is employed to compute fine-grained relevance scores based on the well-selected neighbors. It is worth noting that CDSM is a generic framework which accommodates most of the mainstream graph-based semantic matching networks. The major challenge is how the selector can learn to discriminate the neighbors’ usefulness which has no explicit labels. To cope with this problem, we design a weak-supervision strategy for optimization, where we train the graph-based matching network at first and then the ad-hoc neighbor selector is learned on top of the annotations from the matching network. We conduct extensive experiments with three large-scale datasets, showing that CDSM notably improves the semantic matching accuracy and efficiency thanks to the selection of high-quality neighbors. The source code is released at https://github.com/jingjyyao/CDSM.

User identity linkage, which aims to link identities of a natural person across different social platforms, has attracted increasing research interest recently. Existing approaches usually first embed the identities as deterministic vectors in a shared latent space, and then learn a classifier based on the available annotations. However, the formation and characteristics of real-world social platforms are full of uncertainties, which makes these deterministic embedding based methods sub-optimal. Besides, semi-supervised models utilize the unlabeled data to help capture the intrinsic data distribution. However, the existing semi-supervised linkage methods heavily rely on the heuristically defined similarity measurements to incorporate the innate closeness between labeled and unlabeled samples. Such manually designed assumptions may not be consistent with the actual linkage signals and further introduce the noises. To address the mentioned limitations, in this paper we propose a novel Noise-aware Semi-supervised Variational User Identity Linkage (NSVUIL) model. Specifically, we first propose a novel supervised linkage module to incorporate the available annotations. Each social identity is represented by a Gaussian distribution in the Wasserstein space to simultaneously preserve the fine-grained social profiles and model the uncertainty of identities. Then, a noise-aware self-learning module is designed to faithfully augment the few available annotations, which is capable of filtering noises from the pseudo-labels generated by the supervised module. The filtered reliable candidates are added into the labeled set to provide enhanced training guidance for the next training iteration. Empirically, we evaluate the NSVUIL model over multiple real-world datasets, and the experimental results demonstrate its superiority.

Information network embedding is an important way to enable efficient graph analytics. However, it still faces with computational challenges in problems such as link prediction and node recommendation, particularly with the increasing scale of networks. Both hashing and quantization are promising approaches for accelerating these problems by orders of magnitude. In the preliminary work, we have proposed to learn binary codes for information networks, but graph analytics may suffer from large accuracy degradation. To reduce information loss while achieving memory and search efficiency, we further propose to learn quantized codes for information networks. In particular, each node is represented by compositing multiple latent vectors, each of which is optimally selected from a distinct set. Since (generalized) matrix factorization unifies several well-known embedding methods with high-order proximity preserved, we propose a \underline{N}etwork \underline{R}epresentation \underline{L}ightening framework based on \underline{M}atrix \underline{F}actorization (NRL-MF) to learn binary and quantized codes. We also propose an alternating optimization algorithm for efficient parameter learning, even for the generalized matrix factorization case. We finally evaluate NRL-MF on four real-world information network datasets with respect to the tasks of node classification and node recommendation. The results show that NRL-MF significantly outperforms competing baselines in both tasks, and that quantized representations indeed incur much smaller information loss than binarized codes.

We focus on Maximum Inner Product Search (MIPS), which is an essential problem in many machine learning communities. Given a query, MIPS finds the most similar items with the maximum inner products. Methods for Nearest Neighbor Search (NNS) which is usually defined on metric space do not exhibit the satisfactory performance for MIPS problem since inner product is a non-metric function. However, inner products exhibit many good properties compared with metric functions, such as avoiding vanishing and exploding gradients. As a result, inner product is widely used in many recommendation systems, which makes efficient Maximum Inner Product Search a key for speeding up many recommendation systems. Graph-based methods for NNS problem show the superiorities compared with other class methods. Each data point of the database is mapped to a node of the proximity graph. Nearest neighbor search in the database can be converted to route on the proximity graph to find the nearest neighbor for the query. This technique can be used to solve MIPS problem. Instead of searching the nearest neighbor for the query, we search the item with a maximum inner product with query on the proximity graph. In this article, we propose a reinforcement model to train an agent to search on the proximity graph automatically for MIPS problem if we lack the ground truths of training queries. If we know the ground truths of some training queries, our model can also utilize these ground truths by imitation learning to improve the agent’s searchability. By experiments, we can see that our proposed mode which combines reinforcement learning with imitation learning shows the superiorities over the state-of-the-art methods.

Graph neural networks (GNNs) have been widely applied in the recommendation tasks and have achieved very appealing performance. However, most GNN-based recommendation methods suffer from the problem of data sparsity in practice. Meanwhile, pre-training techniques have achieved great success in mitigating data sparsity in various domains such as natural language processing (NLP) and computer vision (CV) . Thus, graph pre-training has the great potential to alleviate data sparsity in GNN-based recommendations. However, pre-training GNNs for recommendations faces unique challenges. For example, user-item interaction graphs in different recommendation tasks have distinct sets of users and items, and they often present different properties. Therefore, the successful mechanisms commonly used in NLP and CV to transfer knowledge from pre-training tasks to downstream tasks such as sharing learned embeddings or feature extractors are not directly applicable to existing GNN-based recommendations models. To tackle these challenges, we delicately design an adaptive graph pre-training framework for localized collaborative filtering (ADAPT) . It does not require transferring user/item embeddings, and is able to capture both the common knowledge across different graphs and the uniqueness for each graph simultaneously. Extensive experimental results have demonstrated the effectiveness and superiority of ADAPT.

Personalized news recommendation is important for users to find interesting news information and alleviate information overload. Although it has been extensively studied over decades and has achieved notable success in improving user experience, there are still many problems and challenges that need to be further studied. To help researchers master the advances in personalized news recommendation, in this article, we present a comprehensive overview of personalized news recommendation. Instead of following the conventional taxonomy of news recommendation methods, in this article, we propose a novel perspective to understand personalized news recommendation based on its core problems and the associated techniques and challenges. We first review the techniques for tackling each core problem in a personalized news recommender system and the challenges they face. Next, we introduce the public datasets and evaluation methods for personalized news recommendation. We then discuss the key points on improving the responsibility of personalized news recommender systems. Finally, we raise several research directions that are worth investigating in the future. This article can provide up-to-date and comprehensive views on personalized news recommendation. We hope this article can facilitate research on personalized news recommendation as well as related fields in natural language processing and data mining.

The detection of fake news often requires sophisticated reasoning skills, such as logically combining information by considering word-level subtle clues. In this paper, we move towards fine-grained reasoning for fake news detection by better reflecting the logical processes of human thinking and enabling the modeling of subtle clues. In particular, we propose a fine-grained reasoning framework by following the human’s information-processing model, introduce a mutual-reinforcement-based method for incorporating human knowledge about which evidence is more important, and design a prior-aware bi-channel kernel graph network to model subtle differences between pieces of evidence. Extensive experiments show that our model outperforms the state-of-the-art methods and demonstrate the explainability of our approach.

Session-based recommendation (SBR) aims to predict the user's next action based on the ongoing sessions. Recently, there has been an increasing interest in modeling the user preference evolution to capture the fine-grained user interests. While latent user preferences behind the sessions drift continuously over time, most existing approaches still model the temporal session data in discrete state spaces, which are incapable of capturing the fine-grained preference evolution and result in sub-optimal solutions. To this end, we propose Graph Nested GRU ordinary differential equation (ODE), namely GNG-ODE, a novel continuum model that extends the idea of neural ODEs to continuous-time temporal session graphs. The proposed model preserves the continuous nature of dynamic user preferences, encoding both temporal and structural patterns of item transitions into continuous-time dynamic embeddings. As the existing ODE solvers do not consider graph structure change and thus cannot be directly applied to the dynamic graph, we propose a time alignment technique, called t-Alignment, to align the updating time steps of the temporal session graphs within a batch. Empirical results on three benchmark datasets show that GNG-ODE significantly outperforms other baselines.

Candidate generation task requires that candidates related to user interests need to be extracted in realtime. Previous works usually transform a user's behavior sequence to a unified embedding, which can not reflect the user's multiple interests. Some recent works like Comirec and Octopus use multi-channel structures to capture users' diverse interests. They cluster users' historical behaviors into several groups, claiming that one group represents one interest. However, these methods have some limitations. First, an item may correspond to multiple interests of users, thereby simply allocating it to just one interest group will make the modeling of users' interests coarse-grained and inaccurate. Second, explaining user interests at the level of items is rather vague and not convincing. In this paper, we propose a Knowledge Enhanced Multi-Interest Network: KEMI, which exploits knowledge graphs to help learn users' diverse interest representations via heterogeneous graph neural networks (HGNNs) and a novel dual memory network. Specifically, we use HGNNs to capture the semantic representation of knowledge entities and a novel dual memory network to learn a user's diverse interests from his behavior sequence. Through memory slots of the user memory network and the item memory network, we can learn multiple interests for each user and each item. Meanwhile, by binding the entities to the channels of memory networks, we enable it to be explained from the perspective of the knowledge graph, which enhances the interpretability and understanding of user interests. We conduct extensive experiments on two industrial and publicly available datasets. Experimental results demonstrate that our model achieves significant improvements over state-of-the-art baseline models.

Recommender systems play a significant role in online services and have attracted wide attention from both academia and industry. In this paper, we focus on an important, practical, but often overlooked task: domain-level zero-shot recommendation (DZSR). The challenge of DZSR mainly lies in the absence of collaborative behaviors in the target domain, which may be caused by various reasons, such as the domain being newly launched without existing user-item interactions, or users' behaviors being too sensitive to collect for training. To address this challenge, we propose a Transferable Interest Graph Embedding technique for Recommendations (Tiger). The key idea is to connect isolated collaborative filtering datasets with a knowledge graph tailored to recommendations, then propagate collaborative signals from public domains to the zero-shot target domain. The backbone of Tiger is the transferable interest extractor, which is a simple yet effective graph convolutional network (GCN) aggregating multiple hops of neighbors on a shared interest graph. We find that the bottom layers of GCN preserve more domain-specific information while the upper layers represent universal interest better. Thus, in Tiger, we discard the bottom layers of GCN to reconstruct user interest so that collaborative signals can be successfully propagated to other domains, and retain the bottom layers of GCN to include domain-specific information for items. Extensive experiments with four public datasets demonstrate that Tiger can effectively make recommendations for a zero-shot domain and outperform several alternative baselines.

Recently, powerful Transformer architectures have proven superior in generating high-quality sentences.Nevertheless, these models tend to produce dull high-frequency phrases, severely hurting the diversity and novelty of generated text.In this work, we dig into the intrinsic mechanism of this problem and found that sparser attention values in Transformer could improve diversity.To understand such a phenomenon, we first conduct both empirical and theoretical analysis and then attribute it to representation degeneration caused by the attentive mixture of the hidden states during training.We term this process the Trap of Mediocrity.To escape from such a trap, we introduce a novel attention regularization loss to control the sharpness of the attention distribution, which is transparent to model structures and can be easily implemented within 20 lines of python code.We prove that this method could be mathematically regarded as learning a Bayesian approximation of posterior attention.Experiments show that our method improved the diversity and novelty of the generated text while maintaining comparable quality on a variety of conditional and unconditional generation tasks.

The bloom of the Internet and the recent breakthroughs in deep learning techniques open a new door to AI for E-commence, with a trend evolved from using a few financial factors such as liquidity and profitability to using more advanced AI techniques to process complex and multi-modal data.In this paper, we tackle the practical problem of restaurant survival prediction.We argue that traditional methods ignore two essential aspects, which are very helpful for the task: 1) modeling customer reviews and 2) jointly considering status prediction and result explanation.Thus, we propose a novel joint learning framework for explainable restaurant survival prediction based on the multi-modal data of user-restaurant interactions and users' textual reviews.Moreover, we design a graph neural network to capture the high-order interactions and design a co-attention mechanism to capture the most informative and meaningful signal from noisy textual reviews.Our results on two datasets show a significant and consistent improvement over the SOTA techniques (average 6.8% improvement in prediction and 45.3% improvement in explanation).

Zhoujin Tian, Chaozhuo Li, Shuo Ren, Zhiqiang Zuo, Zengxuan Wen, Xinyue Hu, Xiao Han, Haizhen Huang, Denvy Deng, Qi Zhang, Xing Xie. Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing. 2022.

Query-aware webpage snippet extraction is widely used in search engines to help users better understand the content of the returned webpages before clicking.Although important, it is very rarely studied.In this paper, we propose an effective query-aware webpage snippet extraction method named DeepQSE, aiming to select a few sentences which can best summarize the webpage content in the context of input query.DeepQSE first learns query-aware sentence representations for each sentence to capture the fine-grained relevance between query and sentence, and then learns document-aware query-sentence relevance representations for snippet extraction.Since the query and each sentence are jointly modeled in DeepQSE, its online inference may be slow.Thus, we further propose an efficient version of DeepQSE, named Efficient-DeepQSE, which can significantly improve the inference speed of Deep-QSE without affecting its performance.The core idea of Efficient-DeepQSE is to decompose the query-aware snippet extraction task into two stages, i.e., a coarse-grained candidate sentence selection stage where sentence representations can be cached, and a fine-grained relevance modeling stage.Experiments on two real-world datasets validate the effectiveness and efficiency of our methods.

The effectiveness of knowledge graph embedding (KGE) largely depends on the ability to model intrinsic relation patterns and mapping properties. However, existing approaches can only capture some of them with insufficient modeling capacity. In this work, we propose a more powerful KGE framework named HousE, which involves a novel parameterization based on two kinds of Householder transformations: (1) Householder rotations to achieve superior capacity of modeling relation patterns; (2) Householder projections to handle sophisticated relation mapping properties. Theoretically, HousE is capable of modeling crucial relation patterns and mapping properties simultaneously. Besides, HousE is a generalization of existing rotation-based models while extending the rotations to high-dimensional spaces. Empirically, HousE achieves new state-of-the-art performance on five benchmark datasets. Our code is available at https://github.com/anrep/HousE.

Clickbait refers to deliberately created sensational or deceptive text for tricking readers into clicking, which severely hurts the web ecosystem. With a growing number of clickbaits on social media, developing automatic detection methods becomes essential. Nonetheless, the performance of existing neural classifiers is limited due to the underutilization of small labelled datasets. Inspired by related pedagogy theories that learning to write can promote comprehension ability, we propose a novel Contrastive Variational Modelling (CVM) framework to exploit the labelled data better. CVM models the conditional distributions of text and clickbait labels by predicting labels from text and generating text from labels simultaneously with Variational AutoEncoder and further differentiates the learned spaces under each label by a mixed contrastive learning loss. In this way, CVM can capture more underlying textual properties and hence utilize label information to its full potential, boosting detection performance. We theoretically demonstrate CVM as learning a joint distribution of text, clickbait label, and latent variable. Experiments on three clickbait detection datasets show our method's robustness to inadequate and biased labels, outperforming several recent strong baselines.

Chit-chat has been shown effective in engaging users in human-computer interaction. We find with a user study that generating appropriate chit-chat for news articles can help expand user interest and increase the probability that a user reads a recommended news article. Based on this observation, we propose a method to generate personalized chit-chat for news recommendation. Different from existing methods for personalized text generation, our method only requires an external chat corpus obtained from an online forum, which can be disconnected from the recommendation dataset from both the user and item (news) perspectives. This is achieved by designing a weak supervision method for estimating users' personalized interest in a chit-chat post by transferring knowledge learned by a news recommendation model. Based on the method for estimating user interest, a reinforcement learning framework is proposed to generate personalized chit-chat. Extensive experiments, including the automatic offline evaluation and user studies, demonstrate the effectiveness of our method.

Federated learning (FL) is an important paradigm for training global models from decentralized data in a privacy-preserving way. Existing FL methods usually assume the global model can be trained on any participating client. However, in real applications, the devices of clients are usually heterogeneous, and have different computing power. Although big models like BERT have achieved huge success in AI, it is difficult to apply them to heterogeneous FL with weak clients. The straightforward solutions like removing the weak clients or using a small model to fit all clients will lead to some problems, such as under-representation of dropped clients and inferior accuracy due to data loss or limited model representation ability. In this work, we propose InclusiveFL, a client-inclusive federated learning method to handle this problem. The core idea of InclusiveFL is to assign models of different sizes to clients with different computing capabilities, bigger models for powerful clients and smaller ones for weak clients. We also propose an effective method to share the knowledge among local models with different sizes. In this way, all the clients can participate in FL training, and the final model can be big and powerful enough. Besides, we propose a momentum knowledge distillation method to better transfer knowledge in big models on powerful clients to the small models on weak clients. Extensive experiments on many real-world benchmark datasets demonstrate the effectiveness of InclusiveFL in learning accurate models from clients with heterogeneous devices under the FL framework.

As the fundamental basis of sponsored search, relevance modeling measures the closeness between the input queries and the candidate ads. Conventional relevance models solely rely on the textual data, which suffer from the scarce semantic signals within the short queries. Recently, user historical click behaviors are incorporated in the format of click graphs to provide additional correlations beyond pure textual semantics, which contributes to advancing the relevance modeling performance. However, user behaviors are usually arbitrary and unpredictable, leading to the noisy and sparse graph topology. In addition, there exist other types of user behaviors besides clicks, which may also provide complementary information. In this paper, we study the novel problem of heterogeneous behavior graph learning to facilitate relevance modeling task. Our motivation lies in learning an optimal and task-relevant heterogeneous behavior graph consisting of multiple types of user behaviors. We further propose a novel HBGLR model to learn the behavior graph structure by mining the sophisticated correlations between node semantics and graph topology, and encode the textual semantics and structural heterogeneity into the learned representations. Our proposal is evaluated over real-world industry datasets, and has been mainstreamed in the Bing ads. Both offline and online experimental results demonstrate its superiority.

Friend recommendation service plays an important role in shaping and facilitating the growth of online social networks. Graph embedding models, which can learn low-dimensional embeddings for nodes in the social graph to effectively represent the proximity between nodes, have been widely adopted for friend recommendations. Recently, Graph Neural Networks (GNNs) have demonstrated superiority over shallow graph embedding methods, thanks to their ability to explicitly encode neighborhood context. This is also verified in our Xbox friend recommendation scenario, where some simplified GNNs, such as LightGCN and PPRGo, achieve the best performance. However, we observe that many GNN variants, including LightGCN and PPRGo, use a static and pre-defined normalizer in neighborhood aggregation, which is decoupled with the representation learning process and can cause the scale distortion issue. As a consequence, the true power of GNNs has not yet been fully demonstrated in friend recommendations.

Federated learning (FL) is a feasible technique to learn personalized recommendation models from decentralized user data. Unfortunately, federated recommender systems are vulnerable to poisoning attacks by malicious clients. Existing recommender system poisoning methods mainly focus on promoting the recommendation chances of target items due to financial incentives. In fact, in real-world scenarios, the attacker may also attempt to degrade the overall performance of recommender systems. However, existing general FL poisoning methods for degrading model performance are either ineffective or not concealed in poisoning federated recommender systems. In this paper, we propose a simple yet effective and covert poisoning attack method on federated recommendation, named FedAttack. Its core idea is using globally hardest samples to subvert model training. More specifically, the malicious clients first infer user embeddings based on local user profiles. Next, they choose the candidate items that are most relevant to the user embeddings as hardest negative samples, and find the candidates farthest from the user embeddings as hardest positive samples. The model gradients inferred from these poisoned samples are then uploaded for aggregation. Extensive experiments on two benchmark datasets show that FedAttack can effectively degrade the performance of various federated recommender systems, meanwhile cannot be effectively detected nor defended by many existing methods.

News recommendation calls for deep insights of news articles' underlying semantics. Therefore, pretrained language models (PLMs), like BERT and RoBERTa, may substantially contribute to the recommendation quality. However, it's extremely challenging to have news recommenders trained together with such big models: the learning of news recommenders requires intensive news encoding operations, whose cost is prohibitive if PLMs are used as the news encoder. In this paper, we propose a novel framework, SpeedyFeed, which efficiently trains PLMs-based news recommenders of superior quality. SpeedyFeed is highlighted for its light-weight encoding pipeline, which gives rise to three major advantages. Firstly, it makes the intermediate results fully reusable for the training workflow, which removes most of the repetitive but redundant encoding operations. Secondly, it improves the data efficiency of the training workflow, where non-informative data can be eliminated from encoding. Thirdly, it further saves the cost by leveraging simplified news encoding and compact news representation.

The wide spread of fake news has caused serious societal issues. We propose a subgraph reasoning paradigm for fake news detection, which provides a crystal type of explainability by revealing which subgraphs of the news propagation network are the most important for news verification, and concurrently improves the generalization and discrimination power of graph-based detection models by removing task-irrelevant information. In particular, we propose a reinforced subgraph generation method, and perform fine-grained modeling on the generated subgraphs by developing a Hierarchical Path-aware Kernel Graph Attention Network. We also design a curriculum-based optimization method to ensure better convergence and train the two parts in an end-to-end manner.

Embedding based retrieval (EBR) is a fundamental building block in many web applications. However, EBR in sponsored search is distinguished from other generic scenarios and technically challenging due to the need of serving multiple retrieval purposes: firstly, it has to retrieve high-relevance ads, which may exactly serve user's search intent; secondly, it needs to retrieve high-CTR ads so as to maximize the overall user clicks. In this paper, we present a novel representation learning framework Uni-Retriever developed for Bing Search, which unifies two different training modes knowledge distillation and contrastive learning to realize both required objectives. On one hand, the capability of making high-relevance retrieval is established by distilling knowledge from the "relevance teacher model''. On the other hand, the capability of making high-CTR retrieval is optimized by learning to discriminate user's clicked ads from the entire corpus. The two training modes are jointly performed as a multi-objective learning process, such that the ads of high relevance and CTR can be favored by the generated embeddings. Besides the learning strategy, we also elaborate our solution for EBR serving pipeline built upon the substantially optimized DiskANN, where massive-scale EBR can be performed with competitive time and memory efficiency, and accomplished in high-quality. We make comprehensive offline and online experiments to evaluate the proposed techniques, whose findings may provide useful insights for the future development of EBR systems. Uni-Retriever has been mainstreamed as the major retrieval path in Bing's production thanks to the notable improvements on the representation and EBR serving quality.

Jinyi Hu, Xiaoyuan Yi, Wenhao Li, Maosong Sun, Xing Xie. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

We present SELOR, a framework for integrating self-explaining capabilities into a given deep model to achieve both high prediction performance and human precision. By “human precision”, we refer to the degree to which humans agree with the reasons models provide for their predictions. Human precision affects user trust and allows users to collaborate closely with the model. We demonstrate that logic rule explanations naturally satisfy them with the expressive power required for good predictive performance. We then illustrate how to enable a deep model to predict and explain with logic rules. Our method does not require predefined logic rule sets or human annotations and can be learned efficiently and easily with widely-used deep learning modules in a differentiable way. Extensive experiments show that our method gives explanations closer to human decision logic than other methods while maintaining the performance of the deep learning model.

Vertical federated learning (VFL) is a privacy-preserving machine learning paradigm that can learn models from features distributed on different platforms in a privacy-preserving way. Since in real-world applications the data may contain bias on fairness-sensitive features (e.g., gender), VFL models may inherit bias from training data and become unfair for some user groups. However, existing fair machine learning methods usually rely on the centralized storage of fairness-sensitive features to achieve model fairness, which are usually inapplicable in federated scenarios. In this paper, we propose a fair vertical federated learning framework (FairVFL), which can improve the fairness of VFL models. The core idea of FairVFL is to learn unified and fair representations of samples based on the decentralized feature fields in a privacy-preserving way. Specifically, each platform with fairness-insensitive features first learns local data representations from local features. Then, these local representations are uploaded to a server and aggregated into a unified representation for the target task. In order to learn a fair unified representation, we send it to each platform storing fairness-sensitive features and apply adversarial learning to remove bias from the unified representation inherited from the biased data. Moreover, for protecting user privacy, we further propose a contrastive adversarial learning method to remove private information from the unified representation in server before sending it to the platforms keeping fairness-sensitive features. Experiments on three real-world datasets validate that our method can effectively improve model fairness with user privacy well-protected.

Semi-supervised learning (SSL) improves model generalization by leveraging massive unlabeled data to augment limited labeled samples. However, currently, popular SSL evaluation protocols are often constrained to computer vision (CV) tasks. In addition, previous work typically trains deep neural networks from scratch, which is time-consuming and environmentally unfriendly. To address the above issues, we construct a Unified SSL Benchmark (USB) for classification by selecting 15 diverse, challenging, and comprehensive tasks from CV, natural language processing (NLP), and audio processing (Audio), on which we systematically evaluate the dominant SSL methods, and also open-source a modular and extensible codebase for fair evaluation of these SSL methods. We further provide the pre-trained versions of the state-of-the-art neural models for CV tasks to make the cost affordable for further tuning. USB enables the evaluation of a single SSL algorithm on more tasks from multiple domains but with less cost. Specifically, on a single NVIDIA V100, only 39 GPU days are required to evaluate FixMatch on 15 tasks in USB while 335 GPU days (279 GPU days on 4 CV datasets except for ImageNet) are needed on 5 CV tasks with TorchSSL.

Current state-of-the-art document retrieval solutions mainly follow an index-retrieve paradigm, where the index is hard to be directly optimized for the final retrieval target. In this paper, we aim to show that an end-to-end deep neural network unifying training and indexing stages can significantly improve the recall performance of traditional methods. To this end, we propose Neural Corpus Indexer (NCI), a sequence-to-sequence network that generates relevant document identifiers directly for a designated query. To optimize the recall performance of NCI, we invent a prefix-aware weight-adaptive decoder architecture, and leverage tailored techniques including query generation, semantic document identifiers, and consistency-based regularization. Empirical studies demonstrated the superiority of NCI on two commonly used academic benchmarks, achieving +21.4% and +16.8% relative enhancement for Recall@1 on NQ320k dataset and R-Precision on TriviaQA dataset, respectively, compared to the best baseline method.

User-item interactions in recommendations can be naturally denoted as a user-item bipartite graph. Given the success of graph neural networks (GNNs) in graph representation learning, GNN-based Collaborative Filtering (CF) methods have been proposed to advance recommender systems. These methods often make recommendations based on the learned user and item embeddings. However, we found that they do not perform well with sparse user-item graphs which are quite common in real-world recommendations. Therefore, in this work, we introduce a novel perspective to build GNN-based CF methods for recommendations which leads to the proposed framework Localized Graph Collaborative Filtering (LGCF). One key advantage of LGCF is that it does not need to learn embeddings for each user and item, which is challenging in sparse scenarios. Alternatively, LGCF aims at encoding useful CF information into a localized graph and making recommendations based on such graph. Extensive experiments on various datasets validate the effectiveness of LGCF, especially in sparse scenarios. Furthermore, empirical results demonstrate that LGCF provides complementary information to the embedding-based CF model which can be utilized to boost recommendation performance.

A large-scale recommender system usually consists of recall and ranking modules. The goal of ranking modules (aka rankers) is to elaborately discriminate users' preference on item candidates proposed by recall modules. With the success of deep learning techniques in various domains, we have witnessed the mainstream rankers evolve from traditional models to deep neural models. However, the way that we design and use rankers remains unchanged: offline training the model, freezing the parameters, and deploying it for online serving. Actually, the candidate items are determined by specific user requests, in which underlying distributions (e.g., the proportion of items for different categories, the proportion of popular or new items) are highly different from one another in a production environment. The classical parameter-frozen inference manner cannot adapt to dynamic serving circumstances, making rankers' performance compromised.

With the development of deep learning techniques, deep recommendation models also achieve remarkable improvements in terms of recommendation accuracy. However, due to the large number of candidate items in practice and the high cost of preference computation, these methods also suffer from low efficiency of recommendation. The recently proposed tree-based deep recommendation models alleviate the problem by directly learning tree structure and representations under the guidance of recommendation objectives. However, such models have two shortcomings. First, the max-heap assumption in the hierarchical tree, in which the preference for a parent node should be the maximum between the preferences for its children, is difficult to satisfy in their binary classification objectives. Second, the learned index only includes a single tree, which is different from the widely-used multiple trees index, providing an opportunity to improve the accuracy of recommendation.

News recommendation aims to help online news platform users find their preferred news articles. Existing news recommendation methods usually learn models from historical user behaviors on news. However, these behaviors are usually biased on news providers. Models trained on biased user data may capture and even amplify the biases on news providers, and are unfair for some minority news providers. In this paper, we propose a provider fairness-aware news recommendation framework (named ProFairRec), which can learn news recommendation models fair for different news providers from biased user data. The core idea of ProFairRec is to learn provider-fair news representations and provider-fair user representations to achieve provider fairness. To learn provider-fair representations from biased data, we employ provider-biased representations to inherit provider bias from data. Provider-fair and -biased news representations are learned from news content and provider IDs respectively, which are further aggregated to build fair and biased user representations based on user click history. All of these representations are used in model training while only fair representations are used for user-news matching to achieve fair news recommendation. Besides, we propose an adversarial learning task on news provider discrimination to prevent provider-fair news representation from encoding provider bias. We also propose an orthogonal regularization on provider-fair and -biased representations to better reduce provider bias in provider-fair representations. Moreover, ProFairRec is a general framework and can be applied to different news recommendation methods. Extensive experiments on a public dataset verify that our ProFairRec approach can effectively improve the provider fairness of many existing methods and meanwhile maintain their recommendation accuracy.

Vector quantization (VQ) based ANN indexes, such as Inverted File System (IVF) and Product Quantization (PQ), have been widely applied to embedding based document retrieval thanks to the competitive time and memory efficiency. Originally, VQ is learned to minimize the reconstruction loss, i.e., the distortions between the original dense embeddings and the reconstructed embeddings after quantization. Unfortunately, such an objective is inconsistent with the goal of selecting ground-truth documents for the input query, which may cause severe loss of retrieval quality. Recent works identify such a defect, and propose to minimize the retrieval loss through contrastive learning. However, these methods intensively rely on queries with ground-truth documents, whose performance is limited by the insufficiency of labeled data. In this paper, we propose Distill-VQ, which unifies the learning of IVF and PQ within a knowledge distillation framework. In Distill-VQ, the dense embeddings are leveraged as "teachers'', which predict the query's relevance to the sampled documents. The VQ modules are treated as the "students'', which are learned to reproduce the predicted relevance, such that the reconstructed embeddings may fully preserve the retrieval result of the dense embeddings. By doing so, Distill-VQ is able to derive substantial training signals from the massive unlabeled data, which significantly contributes to the retrieval quality. We perform comprehensive explorations for the optimal conduct of knowledge distillation, which may provide useful insights for the learning of VQ based ANN index. We also experimentally show that the labeled data is no longer a necessity for high-quality vector quantization, which indicates Distill-VQ's strong applicability in practice. The evaluations are performed on MS MARCO and Natural Questions benchmarks, where Distill-VQ notably outperforms the SOTA VQ methods in Recall and MRR. Our code is avaliable at https://github.com/staoxiao/LibVQ.

Learning informative representations of users and items from the historical interactions is crucial to collaborative filtering (CF). Existing CF approaches usually model interactions solely within the Euclidean space. However, the sophisticated user-item interactions inherently present highly non-Euclidean anatomy with various types of geometric patterns (i.e., tree-likeness and cyclic structures). The Euclidean-based models may be inadequate to fully uncover the intent factors beneath such hybrid-geometry interactions. To remedy this deficiency, in this paper, we study the novel problem of Geometric Disentangled Collaborative Filtering (GDCF), which aims to reveal and disentangle the latent intent factors across multiple geometric spaces. A novel generative GDCF model is proposed to learn geometric disentangled representations by inferring the high-level concepts associated with user intentions and various geometries. Empirically, our proposal is extensively evaluated over five real-world datasets, and the experimental results demonstrate the superiority of GDCF.

Graph Neural Networks (GNNs) have demonstrated strong power in mining various graph-structure data. Since real-world graphs are usually on a large scale, training scalable GNNs has become one of the research trends in recent years. Existing methods only produce one single model to serve all nodes. However, different nodes may exhibit various properties thus require diverse models, especially when the graph is large. Forcing all nodes to share a unified model will decrease the model's expressiveness. What is worse, some small groups' patterns are prone to be ignored by the model due to their minority, making these nodes unpredictable and even some raising potential unfairness problems. In this paper, we propose a model-agnostic framework Ada-GNN that provides personalized GNN models for specific sets of nodes. Intuitively, it is desirable that every node has its own model. But considering the efficiency and scalability of the framework, we generate specific GNN models at the subgroup-level rather than individual node-level. To be specific, Ada-GNN first splits the original graph into several non-overlapped subgroups and tags each node with its subgroup label. After that, a meta adapter is proposed to adapt a base GNN model to each subgroup rapidly. To better facilitate the global-to-local knowledge adaption, we design a feature enhancement module that captures the distinctions among different subgroups to improve the Ada-GNN's performance. Ada-GNN is model-agnostic and can be equipped to almost all existing scalable GNN based methods such as GraphSAGE, ClusterGCN, SIGN, and SAGN. We conduct extensive experiments with six popular scalable GNN as base methods on two large-scale datasets, and the results consistently demonstrate the generality and superiority of Ada-GNN.

User interest exploration is an important and challenging topic in recommender systems, which alleviates the closed-loop effects between recommendation models and user-item interactions.Contextual bandit (CB) algorithms strive to make a good trade-off between exploration and exploitation so that users' potential interests have chances to expose. However, classical CB algorithms can only be applied to a small, sampled item set (usually hundreds), which forces the typical applications in recommender systems limited to candidate post-ranking, homepage top item ranking, ad creative selection, or online model selection (A/B test). In this paper, we introduce two simple but effective hierarchical CB algorithms to make a classical CB model (such as LinUCB and Thompson Sampling) capable to explore users' interest in the entire item space without limiting to a small item set. We first construct a hierarchy item tree via a bottom-up clustering algorithm to organize items in a coarse-to-fine manner. Then we propose ahierarchical CB (HCB) algorithm to explore users' interest on the hierarchy tree. HCB takes the exploration problem as a series of decision-making processes, where the goal is to find a path from the root to a leaf node, and the feedback will be back-propagated to all the nodes in the path. We further propose aprogressive hierarchical CB (pHCB) algorithm, which progressively extends visible nodes which reach a confidence level for exploration, to avoid misleading actions on upper-level nodes in the sequential decision-making process. Extensive experiments on two public recommendation datasets demonstrate the effectiveness and flexibility of our methods.

Recommender system is playing an increasingly important role in online news platforms nowadays. Recently, there is a growing demand for applying reinforcement learning (RL) algorithms to news recommendation aiming to maximize long-term and/or non-differentiable objectives. However, without an interactive simulated environment, it is extremely costly to develop powerful RL agents for news recommendation. In this paper, we build a user simulator, namely MINDSim, for news recommendation. Targeting at new user generation and corresponding behavior simulation, we first construct a hidden space for users using a generative adversarial network, so that new users can be generated by sampling from this hidden space. To capture complex and fast user interest drifts over time, we adopt an encoder-decoder architecture, which takes the clicked news during the simulation as input and outputs the new user interests for the next period of time. Finally, we build the MINDSim simulator using MIcrosoft News Dataset (MIND), and extensive experimental results on this large-scale real-world dataset demonstrate that MINDSim can simulate the behaviors of real users with high quality.

Knowledge graphs (KGs) have been widely used to improve recommendation accuracy. The multi-hop paths on KGs also enable recommendation reasoning, which is considered a crystal type of explainability. In this paper, we propose a reinforcement learning framework for multi-level recommendation reasoning over KGs, which leverages both ontology-view and instance-view KGs to model multi-level user interests. This framework ensures convergence to a more satisfying solution by effectively transferring high-level knowledge to lower levels. Based on the framework, we propose a multi-level reasoning path extraction method, which automatically selects between high-level concepts and low-level ones to form reasoning paths that better reveal user interests. Experiments on three datasets demonstrate the effectiveness of our method.

Accurate user interest modeling is important for news recommendation. Most existing methods for news recommendation rely on implicit feedbacks like click for inferring user interests and model training. However, click behaviors usually contain heavy noise, and cannot help infer complicated user interest such as dislike. Besides, the feed recommendation models trained solely on click behaviors cannot optimize other objectives such as user engagement. In this paper, we present a news feed recommendation method that can exploit various kinds of user feedbacks to enhance both user interest modeling and model training. We propose a unified user modeling framework to incorporate various explicit and implicit user feedbacks to infer both positive and negative user interests. In addition, we propose a strong-to-weak attention network that uses the representations of stronger feedbacks to distill positive and negative user interests from implicit weak feedbacks for accurate user interest modeling. Besides, we propose a multi-feedback model training framework to learn an engagement-aware feed recommendation model. Extensive experiments on a real-world dataset show that our approach can effectively improve the model performance in terms of both news clicks and user engagement.

Ad-hoc search calls for the selection of appropriate answers from a massive-scale corpus. Nowadays, the embedding-based retrieval (EBR) becomes a promising solution, where deep learning based document representation and ANN search techniques are allied to handle this task. However, a major challenge is that the ANN index can be too large to fit into memory, given the considerable size of answer corpus. In this work, we tackle this problem with Bi-Granular Document Representation, where the lightweight sparse embeddings are indexed and standby in memory for coarse-grained candidate search, and the heavyweight dense embeddings are hosted in disk for fine-grained post verification. For the best of retrieval accuracy, a Progressive Optimization framework is designed. The sparse embeddings are learned ahead for high-quality search of candidates. Conditioned on the candidate distribution induced by the sparse embeddings, the dense embeddings are continuously learned to optimize the discrimination of ground-truth from the shortlisted candidates. Besides, two techniques: the contrastive quantization and the locality-centric sampling are introduced for the learning of sparse and dense embeddings, which substantially contribute to their performances. Thanks to the above features, our method effectively handles massive-scale EBR with strong advantages in accuracy: with up to recall gain on million-scale corpus, and up to recall gain on billion-scale corpus. Besides, Our method is applied to a major sponsored search platform with substantial gains on revenue (), Recall () and CTR (). Our code is available at https://github.com/microsoft/BiDR.

Native ad is a popular type of online advertisement that has similar forms with the native content displayed on websites. Native ad click-through rate (CTR) prediction is useful for improving user experience and platform revenue. However, it is challenging due to the lack of explicit user intent, and user behaviors on the platform with native ads may be insufficient to infer users’ interest in ads. Fortunately, user behaviors exist on many online platforms that can provide complementary information for user-interest mining. Thus, leveraging multi-platform user behaviors is useful for native ad CTR prediction. However, user behaviors are highly privacy-sensitive, and the behavior data on different platforms cannot be directly aggregated due to user privacy concerns and data protection regulations. Existing CTR prediction methods usually require centralized storage of user behavior data for user modeling, which cannot be directly applied to the CTR prediction task with multi-platform user behaviors. In this article, we propose a federated native ad CTR prediction method named FedCTR, which can learn user-interest representations from cross-platform user behaviors in a privacy-preserving way. On each platform a local user model learns user embeddings from the local user behaviors on that platform. The local user embeddings from different platforms are uploaded to a server for aggregation, and the aggregated ones are sent to the ad platform for CTR prediction. Besides, we apply local differential privacy and differential privacy to the local and aggregated user embeddings, respectively, for better privacy protection. Moreover, we propose a federated framework for collaborative model training with distributed models and user behaviors. Extensive experiments on real-world dataset show that FedCTR can effectively leverage multi-platform user behaviors for native ad CTR prediction in a privacy-preserving manner.

To solve the information explosion problem and enhance user experience in various online applications, recommender systems have been developed to model users’ preferences. Although numerous efforts have been made toward more personalized recommendations, recommender systems still suffer from several challenges, such as data sparsity and cold-start problems. In recent years, generating recommendations with the knowledge graph as side information has attracted considerable interest. Such an approach can not only alleviate the above mentioned issues for a more accurate recommendation, but also provide explanations for recommended items. In this paper, we conduct a systematical survey of knowledge graph-based recommender systems. We collect recently published papers in this field, and group them into three categories, i.e., embedding-based methods, connection-based methods, and propagation-based methods. Also, we further subdivide each category according to the characteristics of these approaches. Moreover, we investigate the proposed algorithms by focusing on how the papers utilize the knowledge graph for accurate and explainable recommendation. Finally, we propose several potential research directions in this field.

News recommendation is important for online news services. Existing news recommendation models are usually learned from users' news click behaviors. Usually the behaviors of users with the same sensitive attributes (e.g., genders) have similar patterns and news recommendation models can easily capture these patterns. It may lead to some biases related to sensitive user attributes in the recommendation results, e.g., always recommending sports news to male users, which is unfair since users may not receive diverse news information. In this paper, we propose a fairness-aware news recommendation approach with decomposed adversarial learning and orthogonality regularization, which can alleviate unfairness in news recommendation brought by the biases of sensitive user attributes. In our approach, we propose to decompose the user interest model into two components. One component aims to learn a bias-aware user embedding that captures the bias information on sensitive user attributes, and the other aims to learn a bias-free user embedding that only encodes attribute-independent user interest information for fairness-aware news recommendation. In addition, we propose to apply an attribute prediction task to the bias-aware user embedding to enhance its ability on bias modeling, and we apply adversarial learning to the bias-free user embedding to remove the bias information from it. Moreover, we propose an orthogonality regularization method to encourage the bias-free user embeddings to be orthogonal to the bias-aware one to better distinguish the bias-free user embedding from the bias-aware one. For fairness-aware news ranking, we only use the bias-free user embedding. Extensive experiments on benchmark dataset show that our approach can effectively improve fairness in news recommendation with minor performance loss.

Xiang Ao, Xiting Wang, Ling Luo, Ying Qiao, Qing He, Xing Xie. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Tao Qi, Fangzhao Wu, Chuhan Wu, Peiru Yang, Yang Yu, Xing Xie, Yongfeng Huang. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Product quantization (PQ) is a widely used technique for ad-hoc retrieval.Recent studies propose supervised PQ, where the embedding and quantization models can be jointly trained with supervised learning.However, there is a lack of appropriate formulation of the joint training objective; thus, the improvements over previous non-supervised baselines are limited in reality.In this work, we propose the Matching-oriented Product Quantization (MoPQ), where a novel objective Multinoulli Contrastive Loss (MCL) is formulated.With the minimization of MCL, we are able to maximize the matching probability of query and ground-truth key, which contributes to the optimal retrieval accuracy.Given that the exact computation of MCL is intractable due to the demand of vast contrastive samples, we further propose the Differentiable Cross-device Sampling (DCS), which significantly augments the contrastive samples for precise approximation of MCL.We conduct extensive experimental studies on four realworld datasets, whose results verify the effectiveness of MoPQ.The code is available at https://github.com/microsoft/MoPQ.1. We adopt the Asymmetric Distance Computation (ADC) (Jégou et al., 2011), where only keys need to be quantized.

News recommendation is critical for personalized news access.Most existing news recommendation methods rely on centralized storage of users' historical news click behavior data, which may lead to privacy concerns and hazards.Federated Learning is a privacy-preserving framework for multiple clients to collaboratively train models without sharing their private data.However, the computation and communication cost of directly learning many existing news recommendation models in a federated way are unacceptable for user clients.In this paper, we propose an efficient federated learning framework for privacy-preserving news recommendation.Instead of training and communicating the whole model, we decompose the news recommendation model into a large news model maintained in the server and a light-weight user model shared on both server and clients, where news representations and user model are communicated between server and clients.More specifically, the clients request the user model and news representations from the server, and send their locally computed gradients to the server for aggregation.The server updates its global user model with the aggregated gradients, and further updates its news model to infer updated news representations.Since the local gradients may contain private information, we propose a secure aggregation method to aggregate gradients in a privacy-preserving way.Experiments on two real-world datasets show that our method can reduce the computation and communication cost on clients while keep promising model performance.

We study two user demands that are important during the exploitation of explanations in practice: 1) understanding the overall model behavior faithfully with limited cognitive load and 2) predicting the model behavior accurately on unseen instances. We illustrate that the two user demands correspond to two major sub-processes in the human cognitive process and propose a unified framework to fulfill them simultaneously. Given a local explanation method, our framework jointly 1) learns a limited number of groupwise explanations that interpret the model behavior on most instances with high fidelity and 2) specifies the region where each explanation applies. Experiments on six datasets demonstrate the effectiveness of our method.

Accurate user modeling is critical for news recommendation. Existing news recommendation methods usually model users' interest from their behaviors via sequential or attentive models. However, they cannot model the rich relatedness between user behaviors, which can provide useful contexts of these behaviors for user interest modeling. In this paper, we propose a novel user modeling approach for news recommendation, which models each user as a personalized heterogeneous graph built from user behaviors to better capture the fine-grained behavior relatedness. In addition, in order to learn user interest embedding from the personalized heterogeneous graph, we propose a novel heterogeneous graph pooling method, which can summarize both node features and graph topology, and be aware of the varied characteristics of different types of nodes. Experiments on large-scale benchmark dataset show the proposed methods can effectively improve the performance of user modeling for news recommendation.

News recommendation systems play a key role in online news reading service. Knowledge graphs (KG), which contain comprehensive structural knowledge, are well known for their potential to enhance both accuracy and explainability. While existing works intensively study using KG to improve news recommendation accuracy, using KG for news recommendation reasoning has not been fully explored. A few works such as KPRN [18], [22] and ADAC [25] have discussed knowledge reasoning in some other recommendation domains such as music or movie, but their methods are not practical for the news. How to make reasoning scalable to generic KGs, easy to deploy for real-time serving and meanwhile elastic for both recall and ranking stages remains an open question.

We study how pretrained language models can be enhanced by using deep reinforcement learning to generate attractive text advertisements that reach the high quality standard of real-world advertiser mediums. To improve ad attractiveness without hampering user experience, we propose a model-based reinforcement learning framework for text ad generation, which constructs a model for the environment dynamics and avoids large sample complexity. Based on the framework, we develop Masked-Sequence Policy Gradient, a reinforcement learning algorithm that integrates efficiently with pretrained models and explores the action space effectively. Our method has been deployed to production in Microsoft Bing. Automatic offline experiments, human evaluation, and online experiments demonstrate the superior performance of our method.

The representation learning on textual graph is to generate low-dimensional embeddings for the nodes based on the individual textual features and the neighbourhood information. Recent breakthroughs on pretrained language models and graph neural networks push forward the development of corresponding techniques. The existing works mainly rely on the cascaded model architecture: the textual features of nodes are independently encoded by language models at first; the textual embeddings are aggregated by graph neural networks afterwards. However, the above architecture is limited due to the independent modeling of textual features. In this work, we propose GraphFormers, where layerwise GNN components are nested alongside the transformer blocks of language models. With the proposed architecture, the text encoding and the graph aggregation are fused into an iterative workflow, making each node's semantic accurately comprehended from the global perspective. In addition, a progressive learning strategy is introduced, where the model is successively trained on manipulated data and original data to reinforce its capability of integrating information on graph. Extensive evaluations are conducted on three large-scale benchmark datasets, where GraphFormers outperform the SOTA baselines with comparable running efficiency. The source code is released at https://github.com/microsoft/GraphFormers .

Self-attention networks (SANs) have been intensively applied for sequential recommenders, but they are limited due to: (1) the quadratic complexity and vulnerability to over-parameterization in self-attention; (2) inaccurate modeling of sequential relations between items due to the implicit position encoding. In this work, we propose the low-rank decomposed self-attention networks (LightSANs) to overcome these problems. Particularly, we introduce the low-rank decomposed self-attention, which projects user's historical items into a small constant number of latent interests and leverages item-to-interest interaction to generate the context-aware representation. It scales linearly w.r.t. the user's historical sequence length in terms of time and space, and is more resilient to over-parameterization. Besides, we design the decoupled position encoding, which models the sequential relations between items more precisely. Extensive experimental studies are carried out on three real-world datasets, where LightSANs outperform the existing SANs-based recommenders in terms of both effectiveness and efficiency.

Sponsored search ads appear next to search results when people look for products and services on search engines. In recent years, they have become one of the most lucrative channels for marketing. As the fundamental basis of search ads, relevance modeling has attracted increasing attention due to the significant research challenges and tremendous practical value. Most existing approaches solely rely on the semantic information in the input query-ad pair, while the pure semantic information in the short ads data is not sufficient to fully identify user's search intents. Our motivation lies in incorporating the tremendous amount of unsupervised user behavior data from the historical search logs as the complementary graph to facilitate relevance modeling. In this paper, we extensively investigate how to naturally fuse the semantic textual information with the user behavior graph, and further propose three novel AdsGNN models to aggregate topological neighborhood from the perspectives of nodes, edges and tokens. Furthermore, two critical but rarely investigated problems, domain-specific pre-training and long-tail ads matching, are studied thoroughly. Empirically, we evaluate the AdsGNN models over the large industry dataset, and the experimental results of online/offline tests consistently demonstrate the superiority of our proposal.

Representation learning on user-item graph for recommendation has evolved from using single ID or interaction history to exploiting higher-order neighbors. This leads to the success of graph convolution networks (GCNs) for recommendation such as PinSage and LightGCN. Despite effectiveness, we argue that they suffer from two limitations: (1) high-degree nodes exert larger impact on the representation learning, deteriorating the recommendations of low-degree (long-tail) items; and (2) representations are vulnerable to noisy interactions, as the neighborhood aggregation scheme further enlarges the impact of observed edges.

Graph Neural Networks (GNNs) have drawn considerable attention in recent years and achieved outstanding performance in many tasks. Most empirical studies of GNNs assume that the observed graph represents a complete and accurate picture of node relationship. However, this fundamental assumption cannot always be satisfied, since the real-world graphs from complex systems are error-prone and may not be compatible with the properties of GNNs. Therefore, GNNs solely relying on original graph may cause unsatisfactory results, one typical example of which is that GNNs perform well on graphs with homophily while fail on the disassortative situation. In this paper, we propose graph estimation neural networks GEN, which estimates graph structure for GNNs. Specifically, our GEN presents a structure model to fit the mechanism of GNNs by generating graphs with community structure, and an observation model that injects multifaceted observations into calculating the posterior distribution of graphs and is the first to incorporate multi-order neighborhood information. With above two models, the estimation of graph is implemented based on Bayesian inference to maximize the posterior probability, which attains mutual optimization with GNN parameters in an iterative framework. To comprehensively evaluate the performance of GEN, we perform a set of experiments on several benchmark datasets with different homophily and a synthetic dataset, where the experimental results demonstrate the effectiveness of our GEN and rationality of the estimated graph.

Large-scale events attracting many participants might have a strong negative impact on productivity, mobility, comfort, and safety in a city. Within the few years, serious accidents led by congestion have occurred, especially during sports events, religious ceremonies, festivals, and so on. To alleviate these serious accidents, predicting the occurrence of a large-scale event is much significant. When we know an event occurrence in advance, some of those who are not interested in the event might change their plans and/or might take a detour to avoid to get involved in a heavy congestion. In this paper, we present an early event detection technique named BusBeat. BusBeat uses GPS trajectory data collected from periodic-cars that are vehicles periodically traveling on a pre-scheduled route with a pre-determined departure time, such as a transit bus, shuttle, garbage truck, or municipal patrol car. BusBeat interpolates the missing GPS data by using the features of the periodic-cars. In addition, BusBeat uses the network-based analysis with a Time-dependent Congestion Network (TCN) in order to detect geo-spatial events. BusBeat achieves early event detection without incurring any privacy invasion, by using the continuous trajectories of the periodic-cars that provide the real-time traffic flow and speed. Since traffic towards an event venue would be slow before the event starts, BusBeat detects the geo-spatial events before the attendees gather. We evaluate our BusBeat using over 7,000-bus data collected in Beijing for 5 months and compare with the check-in data collected from a social network service.

The relationships between objects in a network are typically diverse and complex, leading to the heterogeneous edges with different semantic information. In this article, we focus on exploring the heterogeneous edges for network representation learning. By considering each relationship as a view that depicts a specific type of proximity between nodes, we propose a multi-stage non-negative matrix factorization (MNMF) model, committed to utilizing abundant information in multiple views to learn robust network representations. In fact, most existing network embedding methods are closely related to implicitly factorizing the complex proximity matrix. However, the approximation error is usually quite large, since a single low-rank matrix is insufficient to capture the original information. Through a multi-stage matrix factorization process motivated by gradient boosting, our MNMF model achieves lower approximation error. Meanwhile, the multi-stage structure of MNMF gives the feasibility of designing two kinds of non-negative matrix factorization (NMF) manners to preserve network information better. The united NMF aims to preserve the consensus information between different views, and the independent NMF aims to preserve unique information of each view. Concrete experimental results on realistic datasets indicate that our model outperforms three types of baselines in practical applications.

Binary representation of users and items can dramatically improve efficiency of recommendation and reduce size of recommendation models. However, learning optimal binary codes for them is challenging due to binary constraints, even if squared loss is optimized. In this article, we propose a general framework for discrete matrix factorization based on discrete optimization, which can 1) optimize multiple loss functions; 2) handle both explicit and implicit feedback datasets; and 3) take auxiliary information into account without any hyperparameters. To tackle the challenging discrete optimization problem, we propose block coordinate descent based on semidefinite relaxation of binary quadratic programming. We theoretically show that it is equivalent to discrete coordinate descent when only one coordinate is in each block. We extensively evaluate the proposed algorithms on eight real-world datasets. The results of evaluation show that they outperform the state-of-the-art baselines significantly and that auxiliary information of items improves recommendation performance. For better showing the advantages of binary representation, we further propose a two-stage recommender system, consisting of an item-recalling stage and a subsequent fine-ranking stage. Its extensive evaluation shows hashing can dramatically accelerate item recommendation with little degradation of accuracy.

Because of strict response-time constraints, efficiency of top-k recommendation is crucial for real-world recommender systems. Locality sensitive hashing and index-based methods usually store both index data and item feature vectors in main memory, so they handle a limited number of items. Hashing-based recommendation methods enjoy low memory cost and fast retrieval of items, but suffer from large accuracy degradation. In this paper, we propose product Quantized Collaborative Filtering (pQCF) for better trade-off between efficiency and accuracy. pQCF decomposes a joint latent space of users and items into a Cartesian product of low-dimensional subspaces, and learns clustered representation within each subspace. A latent factor is then represented by a short code, which is composed of subspace cluster indexes. A user's preference for an item can be efficiently calculated via table lookup. We then develop block coordinate descent for efficient optimization and reveal the learning of latent factors is seamlessly integrated with quantization. We further investigate an asymmetric pQCF, dubbed as QCF, where user latent factors are not quantized and shared across different subspaces. The extensive experiments with 6 real-world datasets show that pQCF significantly outperforms the state-of-the-art hashing-based CF and QCF increases recommendation accuracy compared to pQCF.

Graph representation learning aims to embed each vertex in a graph into a low-dimensional vector space. Existing graph representation learning methods can be classified into two categories: generative models that learn the underlying connectivity distribution in a graph, and discriminative models that predict the probability of edge between a pair of vertices. In this paper, we propose GraphGAN, an innovative graph representation learning framework unifying the above two classes of methods, in which the generative and the discriminative model play a game-theoretical minimax game. Specifically, for a given vertex, the generative model tries to fit its underlying true connectivity distribution over all other vertices and produces “fake” samples to fool the discriminative model, while the discriminative model tries to detect whether the sampled vertex is from ground truth or generated by the generative model. With the competition between these two models, both of them can alternately and iteratively boost their performance. Moreover, we propose a novel graph softmax as the implementation of the generative model to overcome the limitations of traditional softmax function, which can be proven satisfying desirable properties of normalization, graph structure awareness, and computational efficiency. Through extensive experiments on real-world datasets, we demonstrate that GraphGAN achieves substantial gains in a variety of applications, including graph reconstruction, link prediction, node classification, recommendation, and visualization, over state-of-the-art baselines.

Natural language inference (NLI) task requires an agent to determine the semantic relation between a premise sentence ( p) and a hypothesis sentence ( h), which demands sufficient understanding about sentences semantic. Due to the issues, such as polysemy, ambiguity, as well as fuzziness of sentences, intense sentence understanding is very challenging. To this end, in this article, we introduce the corresponding image of sentences as reference information for enhancing sentence semantic understanding and representing. Specifically, we first propose an image-enhanced multilevel sentence representation net (IEMLRN), that utilizes the image features from pretrained models for enhancing the sentence semantic understanding at different scales, i.e., lexical-level, phrase-level, and sentence-level. The proposed model advances the performance on NLI tasks by leveraging the pretrained global features of images. However, as these pretrained image features are optimized on specific image classification datasets, they may not show the best performance on NLI tasks. Therefore, we further propose to design an adaptive image feature generator that extracts fine-grained image labels from the corresponding sentences. After that, we extend the IEMLRN to multilevel image-enhanced sentence representation net (MIESR) by utilizing not only the coarse-grained pretrained features of images, but also the fine-grained adaptive features of images. Therefore, sentence semantic can be evaluated and enhanced more comprehensively and precisely. Extensive experiments on two benchmark datasets (SNLI, SICK) clearly show our proposed IEMLRN significantly outperform the state-of-the-art baselines, and our proposed MIESR model achieves the best performance by considering not only the text but also images in an adaptive multigranularities way.

With the explosion of news information, personalized news recommendation has become very important for users to quickly find their interested contents. Most existing methods usually learn the representations of users and news from news contents for recommendation. However, they seldom consider high-order connectivity underlying the user-news interactions. Moreover, existing methods failed to disentangle a user’s latent preference factors which cause her clicks on different news. In this paper, we model the user-news interactions as a bipartite graph and propose a novel Graph Neural News Recommendation model with Unsupervised Preference Disentanglement, named GNUD. Our model can encode high-order relationships into user and news representations by information propagation along the graph. Furthermore, the learned representations are disentangled with latent preference factors by a neighborhood routing algorithm, which can enhance expressiveness and interpretability. A preference regularizer is also designed to force each disentangled subspace to independently reflect an isolated preference, improving the quality of the disentangled representations. Experimental results on real-world news datasets demonstrate that our proposed model can effectively improve the performance of news recommendation and outperform state-of-the-art news recommendation methods.

Personalized news recommendation is a critical technology to improve users' online news reading experience.The core of news recommendation is accurate matching between user's interests and candidate news.The same user usually has diverse interests that are reflected in different news she has browsed.Meanwhile, important semantic features of news are implied in text segments of different granularities.Existing studies generally represent each user as a single vector and then match the candidate news vector, which may lose fine-grained information for recommendation.In this paper, we propose FIM, a Finegrained Interest Matching method for neural news recommendation.Instead of aggregating user's all historical browsed news into a unified vector, we hierarchically construct multilevel representations for each news via stacked dilated convolutions.Then we perform finegrained matching between segment pairs of each browsed news and the candidate news at each semantic level.High-order salient signals are then identified by resembling the hierarchy of image recognition for final click prediction.Extensive experiments on a real-world dataset from MSN news validate the effectiveness of our model on news recommendation.

Fangzhao Wu, Ying Qiao, Jiun-Hung Chen, Chuhan Wu, Tao Qi, Jianxun Lian, Danyang Liu, Xing Xie, Jianfeng Gao, Winnie Wu, Ming Zhou. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020.

Recent studies have shown that both accuracy and explainability are important for recommendation. In this paper, we introduce explainable conversational recommendation, which enables incremental improvement of both recommendation accuracy and explanation quality through multi-turn user-model conversation. We show how the problem can be formulated, and design an incremental multi-task learning framework that enables tight collaboration between recommendation prediction, explanation generation, and user feedback integration. We also propose a multi-view feedback integration method to enable effective incremental model update. Empirical results demonstrate that our model not only consistently improves the recommendation accuracy but also generates explanations that fit user interests reflected in the feedbacks.

Sequential location recommendation plays an important role in many applications such as mobility prediction, route planning and location-based advertisements. In spite of evolving from tensor factorization to RNN-based neural networks, existing methods did not make effective use of geographical information and suffered from the sparsity issue. To this end, we propose a Geography-aware sequential recommender based on the Self-Attention Network (GeoSAN for short) for location recommendation. On the one hand, we propose a new loss function based on importance sampling for optimization, to address the sparsity issue by emphasizing the use of informative negative samples. On the other hand, to make better use of geographical information, GeoSAN represents the hierarchical gridding of each GPS point with a self-attention based geography encoder. Moreover, we put forward geography-aware negative samplers to promote the informativeness of negative samples. We evaluate the proposed algorithm with three real-world LBSN datasets, and show that GeoSAN outperforms the state-of-the-art sequential location recommenders by 34.9%. The experimental results further verify significant effectiveness of the new loss function, geography encoder, and geography-aware negative samplers.

Recommendation techniques are important approaches for alleviating information overload. Being often trained on implicit user feedback, many recommenders suffer from the sparsity challenge due to the lack of explicitly negative samples. The GAN-style recommenders (i.e., IRGAN) addresses the challenge by learning a generator and a discriminator adversarially, such that the generator produces increasingly difficult samples for the discriminator to accelerate optimizing the discrimination objective. However, producing samples from the generator is very time-consuming, and our empirical study shows that the discriminator performs poor in top-k item recommendation. To this end, a theoretical analysis is made for the GAN-style algorithms, showing that the generator of limit capacity is diverged from the optimal generator. This may interpret the limitation of discriminator's performance. Based on these findings, we propose a Sampling-Decomposable Generative Adversarial Recommender (SD-GAR). In the framework, the divergence between some generator and the optimum is compensated by self-normalized importance sampling; the efficiency of sample generation is improved with a sampling-decomposable generator, such that each sample can be generated in O(1) with the Vose-Alias method. Interestingly, due to decomposability of sampling, the generator can be optimized with the closed-form solutions in an alternating manner, being different from policy gradient in the GAN-style algorithms. We extensively evaluate the proposed algorithm with five real-world recommendation datasets. The results show that SD-GAR outperforms IRGAN by 12.4% and the SOTA recommender by 10% on average. Moreover, discriminator training can be 20x faster on the dataset with more than 120K items.

News articles usually contain knowledge entities such as celebrities or organizations. Important entities in articles carry key messages and help to understand the content in a more direct way. An industrial news recommender system contains various key applications, such as personalized recommendation, item-to-item recommendation, news category classification, news popularity prediction and local news detection. We find that incorporating knowledge entities for better document understanding benefits these applications consistently. However, existing document understanding models either represent news articles without considering knowledge entities (e.g., BERT) or rely on a specific type of text encoding model (e.g., DKN) so that the generalization ability and efficiency is compromised. In this paper, we propose KRED, which is a fast and effective model to enhance arbitrary document representation with a knowledge graph. KRED first enriches entities’ embeddings by attentively aggregating information from their neighborhood in the knowledge graph. Then a context embedding layer is applied to annotate the dynamic context of different entities such as frequency, category and position. Finally, an information distillation layer aggregates the entity embeddings under the guidance of the original document representation and transforms the document vector into a new one. We advocate to optimize the model with a multi-task framework, so that different news recommendation applications can be united and useful information can be shared across different tasks. Experiments on a real-world Microsoft News dataset demonstrate that KRED greatly benefits a variety of news recommendation applications.

Candidate generation is a critical task for recommendation system, which is technically challenging from two perspectives. On the one hand, recommendation system requires the comprehensive inclusion of user's interested candidates, yet typical deep user modeling approaches would represent each user as an onefold vector, which is hard to capture user's diverse interests. On the other hand, for the sake of practicability, the candidate generation process needs to be both accurate and efficient. Although existing "multi-channel structures'', like memory networks, are more capable of representing user's diverse interests, they may bring in substantial irrelevant candidates and lead to rapid growth of temporal cost. As a result, it remains a tough issue to comprehensively acquire user's interested items in a practical way.

Knowledge graphs have been widely adopted to improve recommendation accuracy. The multi-hop user-item connections on knowledge graphs also endow reasoning about why an item is recommended. However, reasoning on paths is a complex combinatorial optimization problem. Traditional recommendation methods usually adopt brute-force methods to find feasible paths, which results in issues related to convergence and explainability. In this paper, we address these issues by better supervising the path finding process. The key idea is to extract imperfect path demonstrations with minimum labeling efforts and effectively leverage these demonstrations to guide path finding. In particular, we design a demonstration-based knowledge graph reasoning framework for explainable recommendation. We also propose an ADversarial Actor-Critic (ADAC) model for the demonstration-guided path finding. Experiments on three real-world benchmarks show that our method converges more quickly than the state-of-the-art baseline and achieves better recommendation accuracy and explainability.

Deep recommender systems have achieved remarkable improvements in recent years. Despite its superior ranking precision, the running efficiency and memory consumption turn out to be severe bottlenecks in reality. To overcome both limitations, we propose LightRec, a lightweight recommender system which enjoys fast online inference and economic memory consumption. The backbone of LightRec is a total of B codebooks, each of which is composed of W latent vectors, known as codewords. On top of such a structure, LightRec will have an item represented as additive composition of B codewords, which are optimally selected from each of the codebooks. To effectively learn the codebooks from data, we devise an end-to-end learning workflow, where challenges on the inherent differentiability and diversity are conquered by the proposed techniques. In addition, to further improve the representation quality, several distillation strategies are employed, which better preserves user-item relevance scores and relative ranking orders. LightRec is extensively evaluated with four real-world datasets, which gives rise to two empirical findings: 1) compared with those the state-of-the-art lightweight baselines, LightRec achieves over 11% relative improvements in terms of recall performance; 2) compared to conventional recommendation algorithms, LightRec merely incurs negligible accuracy degradation while leads to more than 27x speedup in top-k recommendation.

Chatbots such as Xiaoice have gained huge popularity in recent years. Users frequently mention their favorite works such as songs and movies in conversations with chatbots. Detecting these entities can help design better chat strategies and improve user experience. Existing named entity recognition methods are mainly designed for formal texts, and their performance on the informal chatbot conversation texts may not be optimal. In addition, these methods rely on massive manually annotated data for model training. In this article, we propose a neural approach to detect entities of works for Chinese chatbot. Our approach is based on a language model (LM) long-short term memory (LSTM) convolutional neural network (CNN) conditional random value (CRF), or LM-LSTM-CNN-CRF, framework, which contains a language model to generate context-aware character embeddings, a Bi-LSTM network to learn contextual character representations from global contexts, a CNN to learn character representations from local contexts, and a CRF layer to jointly decode the character label sequence. In addition, we propose an automatic text annotation method via quote marks to reduce the effort of manual annotation. Besides, we propose an iterative data purification method to improve the quality of the automatically constructed labeled data. Massive experiments on a real-world dataset validate that our approach can achieve good performance on entity detection for Chinese chatbots.

Image based social networks are among the most popular social networking services in recent years. With a tremendous amount of images uploaded everyday, understanding users' preferences on user-generated images and making recommendations have become an urgent need. In fact, many hybrid models have been proposed to fuse various kinds of side information (e.g., image visual representation, social network) and user-item historical behavior for enhancing recommendation performance. However, due to the unique characteristics of the user generated images in social image platforms, the previous studies failed to capture the complex aspects that influence users' preferences in a unified framework. Moreover, most of these hybrid models relied on predefined weights in combining different kinds of information, which usually resulted in sub-optimal recommendation performance. To this end, in this paper, we develop a hierarchical attention model for social contextual image recommendation. In addition to basic latent user interest modeling in the popular matrix factorization based recommendation, we identify three key aspects (i.e., upload history, social influence, and owner admiration) that affect each user's latent preferences, where each aspect summarizes a contextual factor from the complex relationships between users and images. After that, we design a hierarchical attention network that naturally mirrors the hierarchical relationship (elements in each aspects level, and the aspect level) of users' latent interests with the identified key aspects. Specifically, by taking embeddings from state-of-the-art deep learning models that are tailored for each kind of data, the hierarchical attention network could learn to attend differently to more or less content. Finally, extensive experimental results on real-world datasets clearly show the superiority of our proposed model.

Recommender systems have been playing an increasingly important role in our daily life due to the explosive growth of information. Accuracy and explainability are two core aspects when we evaluate a recommendation model and have become one of the fundamental trade-offs in machine learning. In this paper, we propose to alleviate the trade-off between accuracy and explainability by developing an explainable deep model that combines the advantages of deep learning-based models and existing explainable methods. The basic idea is to build an initial network based on an explainable deep hierarchy (e.g., Microsoft Concept Graph) and improve the model accuracy by optimizing key variables in the hierarchy (e.g., node importance and relevance). To ensure accurate rating prediction, we propose an attentive multi-view learning framework. The framework enables us to handle sparse and noisy data by co-regularizing among different feature levels and combining predictions attentively. To mine readable explanations from the hierarchy, we formulate personalized explanation generation as a constrained tree node selection problem and propose a dynamic programming algorithm to solve it. Experimental results show that our model outperforms state-of-the-art methods in terms of both accuracy and explainability.

The problem of session-based recommendation aims to predict user actions based on anonymous sessions. Previous methods model a session as a sequence and estimate user representations besides item representations to make recommendations. Though achieved promising results, they are insufficient to obtain accurate user vectors in sessions and neglect complex transitions of items. To obtain accurate item embedding and take complex transitions of items into account, we propose a novel method, i.e. Session-based Recommendation with Graph Neural Networks, SR-GNN for brevity. In the proposed method, session sequences are modeled as graphstructured data. Based on the session graph, GNN can capture complex transitions of items, which are difficult to be revealed by previous conventional sequential methods. Each session is then represented as the composition of the global preference and the current interest of that session using an attention network. Extensive experiments conducted on two real datasets show that SR-GNN evidently outperforms the state-of-the-art session-based recommendation methods consistently.

Sentence semantic matching requires an agent to determine the semantic relation between two sentences, which is widely used in various natural language tasks such as Natural Language Inference (NLI) and Paraphrase Identification (PI). Among all matching methods, attention mechanism plays an important role in capturing the semantic relations and properly aligning the elements of two sentences. Previous methods utilized attention mechanism to select important parts of sentences at one time. However, the important parts of the sentence during semantic matching are dynamically changing with the degree of sentence understanding. Selecting the important parts at one time may be insufficient for semantic understanding. To this end, we propose a Dynamic Re-read Network (DRr-Net) approach for sentence semantic matching, which is able to pay close attention to a small region of sentences at each step and re-read the important words for better sentence semantic understanding. To be specific, we first employ Attention Stack-GRU (ASG) unit to model the original sentence repeatedly and preserve all the information from bottom-most word embedding input to up-most recurrent output. Second, we utilize Dynamic Re-read (DRr) unit to pay close attention to one important word at one time with the consideration of learned information and re-read the important words for better sentence semantic understanding. Extensive experiments on three sentence matching benchmark datasets demonstrate that DRr-Net has the ability to model sentence semantic more precisely and significantly improve the performance of sentence semantic matching. In addition, it is very interesting that some of finding in our experiments are consistent with the findings of psychological research.

Personalized news recommendation is important to help users find their interested news and improve reading experience.A key problem in news recommendation is learning accurate user representations to capture their interests.Users usually have both long-term preferences and short-term interests.However, existing news recommendation methods usually learn single representations of users, which may be insufficient.In this paper, we propose a neural news recommendation approach which can learn both long-and short-term user representations.The core of our approach is a news encoder and a user encoder.In the news encoder, we learn representations of news from their titles and topic categories, and use attention network to select important words.In the user encoder, we propose to learn long-term user representations from the embeddings of their IDs.In addition, we propose to learn short-term user representations from their recently browsed news via GRU network.Besides, we propose two methods to combine long-term and short-term user representations.The first one is using the long-term user representation to initialize the hidden state of the GRU network in short-term user representation.The second one is concatenating both long-and short-term user representations as a unified user vector.Extensive experiments on a real-world dataset show our approach can effectively improve the performance of neural news recommendation.

Aspect term extraction (ATE) aims at identifying all aspect terms in a sentence and is usually modeled as a sequence labeling problem.However, sequence labeling based methods cannot make full use of the overall meaning of the whole sentence and have the limitation in processing dependencies between labels.To tackle these problems, we first explore to formalize ATE as a sequence-tosequence (Seq2Seq) learning task where the source sequence and target sequence are composed of words and labels respectively.At the same time, to make Seq2Seq learning suit to ATE where labels correspond to words one by one, we design the gated unit networks to incorporate corresponding word representation into the decoder, and position-aware attention to pay more attention to the adjacent words of a target word.The experimental results on two datasets show that Seq2Seq learning is effective in ATE accompanied with our proposed gated unit networks and position-aware attention mechanism.

News recommendation can help users find interested news and alleviate information overload.The topic information of news is critical for learning accurate news and user representations for news recommendation.However, it is not considered in many existing news recommendation methods.In this paper, we propose a neural news recommendation approach with topic-aware news representations.The core of our approach is a topic-aware news encoder and a user encoder.In the news encoder we learn representations of news from their titles via CNN networks and apply attention networks to select important words.In addition, we propose to learn topic-aware news representations by jointly training the news encoder with an auxiliary topic classification task.In the user encoder we learn the representations of users from their browsed news and use attention networks to select informative news for user representation learning.Extensive experiments on a real-world dataset validate the effectiveness of our approach.

Estimating in-hospital costs from medical records is an important task with many applications such as accountable care. Existing methods for this task usually rely on manual feature engineering which needs massive domain knowledge, and do not exploit the textual information in medical records, e.g., diagnosis and operation texts. In this paper, we propose a neural in-hospital cost estimation (NICE) approach to estimate the in-hospital costs of patients from their admission records. Our approach can exploit the heterogeneous information in records, such as patient features, diagnosis/operation texts, and the diagnosis/operation IDs, via a multi-view learning framework. In addition, since different words, diagnoses and operations have different importance for cost estimation, we propose a hierarchical attention network to select important words, diagnoses and operations for learning informative record representations. Extensive experiments on a real-world medical dataset validate the effectiveness of our approach.

Sentiment classification is an important task in the sentiment analysis field. Many deep learning based sentiment classification methods have been proposed in recent years. However, these methods usually rely on massive labeled texts to train sentiment classifiers, which are expensive and time-consuming to annotate. Luckily, many high-quality sentiment lexicons have been constructed and can cover a large number of sentiment words. Since sentiment words are the basic units to convey sentiments in texts, these sentiment lexicons have the potential to improve the performance of neural sentiment classification. In this paper, we propose two approaches to exploit sentiment lexicons to enhance neural sentiment classification. In our first approach we use sentiment lexicons to learn sentiment-aware attentions. We propose a word sentiment classification task to classify the sentiments of words in a sentence based on their hidden representations in the attention network of neural sentiment classification models. We jointly train this task with neural sentiment classifier to facilitate the attention network to recognize and highlight sentiment-bearing words. In our second approach we use sentiment lexicons to learn sentiment-aware word embeddings. We design an auxiliary task to classify the sentiments of words in sentiment lexicons based on their word embeddings, and jointly train this task with neural sentiment classifier to encode sentiment information in sentiment lexicons to word embeddings. Extensive experiments on three benchmark datasets validate the effectiveness of our approach.

Review rating prediction is an important task in data mining and natural language processing fields, and has wide applications. Users usually express opinions towards many aspects in their reviews, and the overall review rating is a synthesis of these opinions. However, most existing review rating prediction methods ignore users' opinions on aspects, which is insufficient. In this paper, we propose a neural aspect-aware rating prediction approach for Chinese reviews. In our approach we propose a collaborative learning framework to jointly train review-level rating predictor and multiple aspect-level rating predictors. In our framework different rating predictors share the same review encoder model to exploit the inherent relatedness between them, but have different attention networks to focus on different informative texts for each task. The final review representation for rating prediction is a concatenation of the review representations from all predictors. Since word segmentation of Chinese reviews is usually inaccurate, we propose a multi-view learning model to learn review representations from both words and characters. Extensive experiments on real-world dataset validate the effectiveness of our approach.

Demographics of social media users such as gender are very important for personalized online services. However, the gender information of many users is usually not available. Luckily, the messages posted by social media users can provide rich clues for inferring their genders, since male and female users usually have differences in their message content. In addition, users with different genders often have different patterns in expressing emotions. In this paper, we propose a neural approach for gender prediction in social media based on both content and emotion of messages posted by users. The core of our approach is an emotion-aware hierarchical user representation model. Our model first learns message representations from words using message encoder and then learns user representations from messages using user encoder with hierarchical attention networks selecting important words and messages to learn informative user representations. In addition, we propose two methods to incorporate emotion information in messages into user representation learning. The first one is to incorporate emotion-aware message representations generated by a pre-trained emotion classifier into message representations. The second one is to train emotion-aware message encoders via jointly training our model with an auxiliary emotion classification task. Extensive experiments on two real-world datasets validate the effectiveness of our approach.

Recent years have witnessed a phenomenal success of agent-initiated social e-commerce models, which encourage users to become selling agents to promote items through their social connections. The complex interactions in this type of social e-commerce can be formulated as Heterogeneous Information Networks (HIN), where there are numerous types of relations between three types of nodes, i.e., users, selling agents and items. Learning high quality node embeddings is of key interest, and Graph Convolutional Networks (GCNs) have recently been established as the latest state-of-the-art methods in representation learning. However, prior GCN models have fundamental limitations in both modeling heterogeneous relations and efficiently sampling relevant receptive field from vast neighborhood. To address these problems, we propose RecoGCN, which stands for a RElation-aware CO-attentive GCN model, to effectively aggregate heterogeneous features in a HIN. It makes up current GCN's limitation in modelling heterogeneous relations with a relation-aware aggregator, and leverages the semantic-aware meta-paths to carve out concise and relevant receptive fields for each node. To effectively fuse the embeddings learned from different meta-paths, we further develop a co-attentive mechanism to dynamically assign importance weights to different meta-paths by attending the three-way interactions among users, selling agents and items. Extensive experiments on a real-world dataset demonstrate RecoGCN is able to learn meaningful node embeddings in HIN, and consistently outperforms baseline methods in recommendation tasks.

Neural network methods have achieved great success in sentiment classification. Recent studies have found that incorporating user and product information can effectively improve the performance of review sentiment classification. However, most of these studies only concentrate on the influence of users and products, ignoring the inherent correlation between users or products. This information is important for users or products since they can obtain more information from similar users or products. In this paper, we propose a novel framework for review rating prediction with user and product memory. First, besides the original user or product representations, we construct inferred representations from representative users or products which are stored in memory slots. These memory units can be viewed as refined knowledge representations of users or products learned from the data. Then, we employ two hierarchical networks with user attention and product attention using both the original and inferred representations. Experiments on benchmark datasets show that our method can achieve state-of-the-art performance. Besides, our approach performs much more better in cold-start scenarios where the training data is scarce.

Chuhan Wu, Fangzhao Wu, Mingxiao An, Tao Qi, Jianqiang Huang, Yongfeng Huang, Xing Xie. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Chuhan Wu, Fangzhao Wu, Suyu Ge, Tao Qi, Yongfeng Huang, Xing Xie. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Chuhan Wu, Fangzhao Wu, Tao Qi, Suyu Ge, Yongfeng Huang, Xing Xie. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Diagnosis prediction, which aims to predict future health information of patients from historical electronic health records (EHRs), is a core research task in personalized healthcare. Although some RNN-based methods have been proposed to model sequential EHR data, these methods have two major issues. First, they cannot capture fine-grained progression patterns of patient health conditions. Second, they do not consider the mutual effect between important context (e.g., patient demographics) and historical diagnosis. To tackle these challenges, we propose a model called Co-Attention Memory networks for diagnosis Prediction (CAMP), which tightly integrates historical records, fine-grained patient conditions, and demographics with a three-way interaction architecture built on co-attention. Our model augments RNNs with a memory network to enrich the representation capacity. The memory network enables analysis of fine-grained patient conditions by explicitly incorporating a taxonomy of diseases into an array of memory slots. We instantiate the READ/WRITE operations of the memory network so that the memory cooperates effectively with the patient demographics through co-attention mechanism. Experiments on real-world datasets demonstrate that CAMP consistently performs better than state-of-the-art methods.

We define a unified information-based measure to provide quantitative explanations on how intermediate layers of deep Natural Language Processing (NLP) models leverage information of input words. Our method advances existing explanation methods by addressing issues in coherency and generality. Explanations generated by using our method are consistent and faithful across different timestamps, layers, and models. We show how our method can be applied to four widely used models in NLP and explain their performances on three real-world benchmark datasets.

Despite widespread adoption, recommender systems remain mostly black boxes. Recently, providing explanations about why items are recommended has attracted increasing attention due to its capability to enhance user trust and satisfaction. In this paper, we propose a co-attentive multi-task learning model for explainable recommendation. Our model improves both prediction accuracy and explainability of recommendation by fully exploiting the correlations between the recommendation task and the explanation task. In particular, we design an encoder-selector-decoder architecture inspired by human's information-processing model in cognitive psychology. We also propose a hierarchical co-attentive selector to effectively model the cross knowledge transferred for both tasks. Our model not only enhances prediction accuracy of the recommendation task, but also generates linguistic explanations that are fluent, useful, and highly personalized. Experiments on three public datasets demonstrate the effectiveness of our model.

Deep learning techniques have been widely applied to modern recommendation systems, bringing in flexible and effective ways of user representation. Conventionally, user representations are generated purely in the offline stage. Without referencing to the specific candidate item for recommendation, it is difficult to fully capture user preference from the perspective of interest. More recent algorithms tend to generate user representation at runtime, where user's historical behaviors are attentively summarized w.r.t. the presented candidate item. In spite of the improved efficacy, it is too expensive for many real-world scenarios because of the repetitive access to user's entire history. In this work, a novel user representation framework, Hi-Fi Ark, is proposed. With Hi-Fi Ark, user history is summarized into highly compact and complementary vectors in the offline stage, known as archives. Meanwhile, user preference towards a specific candidate item can be precisely captured via the attentive aggregation of such archives. As a result, both deployment feasibility and superior recommendation efficacy are achieved by Hi-Fi Ark. The effectiveness of Hi-Fi Ark is empirically validated on three real-world datasets, where remarkable and consistent improvements are made over a variety of well-recognized baseline methods.

When recommending or advertising items to users, an emerging trend is to present each multimedia item with a key frame image (e.g., the poster of a movie). As each multimedia item can be represented as multiple fine-grained visual images (e.g., related images of the movie), personalized key frame recommendation is necessary in these applications to attract users' unique visual preferences. However, previous personalized key frame recommendation models relied on users' fine grained image behavior of multimedia items (e.g., user-image interaction behavior), which is often not available in real scenarios. In this paper, we study the general problem of joint multimedia item and key frame recommendation in the absence of the fine-grained user-image behavior. We argue that the key challenge of this problem lies in discovering users' visual profiles for key frame recommendation, as most recommendation models would fail without any users' fine-grained image behavior. To tackle this challenge, we leverage users' item behavior by projecting users(items) in two latent spaces: a collaborative latent space and a visual latent space. We further design a model to discern both the collaborative and visual dimensions of users, and model how users make decisive item preferences from these two spaces. As a result, the learned user visual profiles could be directly applied for key frame recommendation. Finally, experimental results on a real-world dataset clearly show the effectiveness of our proposed model on the two recommendation tasks.

Personalized news recommendation is very important for online news platforms to help users find interested news and improve user experience. News and user representation learning is critical for news recommendation. Existing news recommendation methods usually learn these representations based on single news information, e.g., title, which may be insufficient. In this paper we propose a neural news recommendation approach which can learn informative representations of users and news by exploiting different kinds of news information. The core of our approach is a news encoder and a user encoder. In the news encoder we propose an attentive multi-view learning model to learn unified news representations from titles, bodies and topic categories by regarding them as different views of news. In addition, we apply both word-level and view-level attention mechanism to news encoder to select important words and views for learning informative news representations. In the user encoder we learn the representations of users based on their browsed news and apply attention mechanism to select informative news for user representation learning. Extensive experiments on a real-world dataset show our approach can effectively improve the performance of news recommendation.

User modeling is an essential task for online recommender systems. In the past few decades, collaborative filtering (CF) techniques have been well studied to model users' long term preferences. Recently, recurrent neural networks (RNN) have shown a great advantage in modeling users' short term preference. A natural way to improve the recommender is to combine both long-term and short-term modeling. Previous approaches neglect the importance of dynamically integrating these two user modeling paradigms. Moreover, users' behaviors are much more complex than sentences in language modeling or images in visual computing, thus the classical structures of RNN such as Long Short-Term Memory (LSTM) need to be upgraded for better user modeling. In this paper, we improve the traditional RNN structure by proposing a time-aware controller and a content-aware controller, so that contextual information can be well considered to control the state transition. We further propose an attention-based framework to combine users' long-term and short-term preferences, thus users' representation can be generated adaptively according to the specific context. We conduct extensive experiments on both public and industrial datasets. The results demonstrate that our proposed method outperforms several state-of-art methods consistently.

The success of recommender systems in modern online platforms is inseparable from the accurate capture of users' personal tastes. In everyday life, large amounts of user feedback data are created along with user-item online interactions in a variety of ways, such as browsing, purchasing, and sharing. These multiple types of user feedback provide us with tremendous opportunities to detect individuals' fine-grained preferences. Different from most existing recommender systems that rely on a single type of feedback, we advocate incorporating multiple types of user-item interactions for better recommendations. Based on the observation that the underlying spectrum of user preferences is reflected in various types of interactions with items and can be uncovered by latent relational learning in metric space, we propose a unified neural learning framework, named Multi-Relational Memory Network (MRMN). It can not only model fine-grained user-item relations but also enable us to discriminate between feedback types in terms of the strength and diversity of user preferences. Extensive experiments show that the proposed MRMN model outperforms competitive state-of-the-art algorithms in a wide range of scenarios, including e-commerce, local services, and job recommendations.

News recommendation is very important to help users find interested news and alleviate information overload. Different users usually have different interests and the same user may have various interests. Thus, different users may click the same news article with attention on different aspects. In this paper, we propose a neural news recommendation model with personalized attention (NPA). The core of our approach is a news representation model and a user representation model. In the news representation model we use a CNN network to learn hidden representations of news articles based on their titles. In the user representation model we learn the representations of users based on the representations of their clicked news articles. Since different words and different news articles may have different informativeness for representing news and users, we propose to apply both word- and news-level attention mechanism to help our model attend to important words and news articles. In addition, the same news article and the same word may have different informativeness for different users. Thus, we propose a personalized attention network which exploits the embedding of user ID to generate the query vector for the word- and news-level attentions. Extensive experiments are conducted on a real-world news recommendation dataset collected from MSN news, and the results validate the effectiveness of our approach on news recommendation.

Existing review-based recommendation methods usually use the same model to learn the representations of all users/items from reviews posted by users towards items. However, different users have different preference and different items have different characteristics. Thus, the same word or the similar reviews may have different informativeness for different users and items. In this paper we propose a neural recommendation approach with personalized attention to learn personalized representations of users and items from reviews. We use a review encoder to learn representations of reviews from words, and a user/item encoder to learn representations of users or items from reviews. We propose a personalized attention model, and apply it to both review and user/item encoders to select different important words and reviews for different users/items. Experiments on five datasets validate our approach can effectively improve the performance of neural recommendation.

Demographics of online users such as age and gender play an important role in personalized web applications. However, it is difficult to directly obtain the demographic information of online users. Luckily, search queries can cover many online users and the search queries from users with different demographics usually have some difference in contents and writing styles. Thus, search queries can provide useful clues for demographic prediction. In this paper, we study predicting users' demographics based on their search queries, and propose a neural approach for this task. Since search queries can be very noisy and many of them are not useful, instead of combining all queries together for user representation, in our approach we propose a hierarchical user representation with attention (HURA) model to learn informative user representations from their search queries. Our HURA model first learns representations for search queries from words using a word encoder, which consists of a CNN network and a word-level attention network to select important words. Then we learn representations of users based on the representations of their search queries using a query encoder, which contains a CNN network to capture the local contexts of search queries and a query-level attention network to select informative search queries for demographic prediction. Experiments on two real-world datasets validate that our approach can effectively improve the performance of search query based age and gender prediction and consistently outperform many baseline methods.

Twitter is a popular social media platform for information sharing and dissemination. Many Twitter users post tweets to share their experiences about drugs and adverse drug reactions. Automatic detection of tweets mentioning drug names and adverse drug reactions at a large scale has important applications such as pharmacovigilance. However, detecting drug name and adverse drug reaction mentioning tweets is very challenging, because tweets are usually very noisy and informal, and there are massive misspellings and user-created abbreviations for these mentions. In addition, these mentions are usually context dependent. In this paper, we propose a neural approach with hierarchical tweet representation and multi-head self-attention mechanism to jointly detect tweets mentioning drug names and adverse drug reactions. In order to alleviate the influence of massive misspellings and user-created abbreviations in tweets, we propose to use a hierarchical tweet representation model to first learn word representations from characters and then learn tweet representations from words. In addition, we propose to use multi-head self-attention mechanism to capture the interactions between words to fully model the contexts of tweets. Besides, we use additive attention mechanism to select the informative words to learn more informative tweet representations. Experimental results validate the effectiveness of our approach.

Chinese word segmentation (CWS) is very important for Chinese text processing. Existing methods for CWS usually rely on a large number of labeled sentences to train word segmentation models, which are expensive and time-consuming to annotate. Luckily, the unlabeled data is usually easy to collect and many high-quality Chinese lexicons are off-the-shelf, both of which can provide useful information for CWS. In this paper, we propose a neural approach for Chinese word segmentation which can exploit both lexicon and unlabeled data. Our approach is based on a variant of posterior regularization algorithm, and the unlabeled data and lexicon are incorporated into model training as indirect supervision by regularizing the prediction space of CWS models. Extensive experiments on multiple benchmark datasets in both in-domain and cross-domain scenarios validate the effectiveness of our approach.

To alleviate sparsity and cold start problem of collaborative filtering based recommender systems, researchers and engineers usually collect attributes of users and items, and design delicate algorithms to exploit these additional information. In general, the attributes are not isolated but connected with each other, which forms a knowledge graph (KG). In this paper, we propose Knowledge Graph Convolutional Networks (KGCN), an end-to-end framework that captures inter-item relatedness effectively by mining their associated attributes on the KG. To automatically discover both high-order structure information and semantic information of the KG, we sample from the neighbors for each entity in the KG as their receptive field, then combine neighborhood information with bias when calculating the representation of a given entity. The receptive field can be extended to multiple hops away to model high-order proximity information and capture users' potential long-distance interests. Moreover, we implement the proposed KGCN in a minibatch fashion, which enables our model to operate on large datasets and KGs. We apply the proposed model to three datasets about movie, book, and music recommendation, and experiment results demonstrate that our approach outperforms strong recommender baselines.

Collaborative filtering often suffers from sparsity and cold start problems in real recommendation scenarios, therefore, researchers and engineers usually use side information to address the issues and improve the performance of recommender systems. In this paper, we consider knowledge graphs as the source of side information. We propose MKR, a Multi-task feature learning approach for Knowledge graph enhanced Recommendation. MKR is a deep end-to-end framework that utilizes knowledge graph embedding task to assist recommendation task. The two tasks are associated by crosscompress units, which automatically share latent features and learn high-order interactions between items in recommender systems and entities in the knowledge graph. We prove that crosscompress units have sufficient capability of polynomial approximation, and show that MKR is a generalized framework over several representative methods of recommender systems and multi-task learning. Through extensive experiments on real-world datasets, we demonstrate that MKR achieves substantial gains in movie, book, music, and news recommendation, over state-of-the-art baselines. MKR is also shown to be able to maintain satisfactory performance even if user-item interactions are sparse.

Chinese named entity recognition (CNER) is an important task in Chinese natural language processing field. However, CNER is very challenging since Chinese entity names are highly context-dependent. In addition, Chinese texts lack delimiters to separate words, making it difficult to identify the boundary of entities. Besides, the training data for CNER in many domains is usually insufficient, and annotating enough training data for CNER is very expensive and time-consuming. In this paper, we propose a neural approach for CNER. First, we introduce a CNN-LSTM-CRF neural architecture to capture both local and long-distance contexts for CNER. Second, we propose a unified framework to jointly train CNER and word segmentation models in order to enhance the ability of CNER model in identifying entity boundaries. Third, we introduce an automatic method to generate pseudo labeled samples from existing labeled data which can enrich the training data. Experiments on two benchmark datasets show that our approach can effectively improve the performance of Chinese named entity recognition, especially when training data is insufficient.

Personalized recommendation has received a lot of attention as a highly practical research topic. However, existing recommender systems provide the recommendations with a generic statement such as “Customers who bought this item also bought…”. Explainable recommendation, which makes a user aware of why such items are recommended, is in demand. The goal of our research is to make the users feel as if they are receiving recommendations from their friends. To this end, we formulate a new challenging problem called personalized reason generation for explainable recommendation for songs in conversation applications and propose a solution that generates a natural language explanation of the reason for recommending a song to that particular user. For example, if the user is a student, our method can generate an output such as “Campus radio plays this song at noon every day, and I think it sounds wonderful,” which the student may find easy to relate to. In the offline experiments, through manual assessments, the gain of our method is statistically significant on the relevance to songs and personalization to users comparing with baselines. Large-scale online experiments show that our method outperforms manually selected reasons by 8.2% in terms of click-through rate. Evaluation results indicate that our generated reasons are relevant to songs and personalized to users, and they attract users to click the recommendations.

To address the sparsity and cold-start problem of collaborative filtering, researchers usually make use of side information, such as social networks or item attributes, to improve the performance of recommendation. In this article, we consider the knowledge graph (KG) as the source of side information. To address the limitations of existing embedding-based and path-based methods for KG-aware recommendation, we propose RippleNet , an end-to-end framework that naturally incorporates the KG into recommender systems. RippleNet has two versions: (1) The outward propagation version, which is analogous to the actual ripples on water, stimulates the propagation of user preferences over the set of knowledge entities by automatically and iteratively extending a user’s potential interests along links in the KG. The multiple “ripples” activated by a user’s historically clicked items are thus superposed to form the preference distribution of the user with respect to a candidate item. (2) The inward aggregation version aggregates and incorporates the neighborhood information biasedly when computing the representation of a given entity. The neighborhood can be extended to multiple hops away to model high-order proximity and capture users’ long-distance interests. In addition, we intuitively demonstrate how a KG assists with recommender systems in RippleNet, and we also find that RippleNet provides a new perspective of explainability for the recommended results in terms of the KG. Through extensive experiments on real-world datasets, we demonstrate that both versions of RippleNet achieve substantial gains in a variety of scenarios, including movie, book, and news recommendations, over several state-of-the-art baselines.

Driven by the wave of urbanization in recent decades, the research topic about migrant behavior analysis draws great attention from both academia and the government. Nevertheless, subject to the cost of data collection and the lack of modeling methods, most of existing studies use only questionnaire surveys with sparse samples and non-individual level statistical data to achieve coarse-grained studies of migrant behaviors. In this paper, a partially supervised cross-domain deep learning model named CD-CNN is proposed for migrant/native recognition using mobile phone signaling data as behavioral features and questionnaire survey data as incomplete labels. Specifically, CD-CNN features in decomposing the mobile data into location domain and communication domain, and adopts a joint learning framework that combines two convolutional neural networks with a feature balancing scheme. Moreover, CD-CNN employs a three-step algorithm for training, in which the co-training step is of great value to partially supervised cross-domain learning. Comparative experiments on the city Wuxi demonstrate the high predictive power of CD-CNN. Two interesting applications further highlight the ability of CD-CNN for in-depth migrant behavioral analysis.

The goal of graph representation learning is to embed each vertex in a graph into a low-dimensional vector space. Existing graph representation learning methods can be classified into two categories: generative models that learn the underlying connectivity distribution in the graph, and discriminative models that predict the probability of edge existence between a pair of vertices. In this paper, we propose GraphGAN, an innovative graph representation learning framework unifying above two classes of methods, in which the generative model and discriminative model play a game-theoretical minimax game. Specifically, for a given vertex, the generative model tries to fit its underlying true connectivity distribution over all other vertices and produces "fake" samples to fool the discriminative model, while the discriminative model tries to detect whether the sampled vertex is from ground truth or generated by the generative model. With the competition between these two models, both of them can alternately and iteratively boost their performance. Moreover, when considering the implementation of generative model, we propose a novel graph softmax to overcome the limitations of traditional softmax function, which can be proven satisfying desirable properties of normalization, graph structure awareness, and computational efficiency. Through extensive experiments on real-world datasets, we demonstrate that GraphGAN achieves substantial gains in a variety of applications, including link prediction, node classification, and recommendation, over state-of-the-art baselines.

To address the sparsity and cold start problem of collaborative filtering, researchers usually make use of side information, such as social networks or item attributes, to improve recommendation performance. This paper considers the knowledge graph as the source of side information. To address the limitations of existing embedding-based and path-based methods for knowledge-graph-aware recommendation, we propose RippleNet, an end-to-end framework that naturally incorporates the knowledge graph into recommender systems. Similar to actual ripples propagating on the water, RippleNet stimulates the propagation of user preferences over the set of knowledge entities by automatically and iteratively extending a user's potential interests along links in the knowledge graph. The multiple "ripples" activated by a user's historically clicked items are thus superposed to form the preference distribution of the user with respect to a candidate item, which could be used for predicting the final clicking probability. Through extensive experiments on real-world datasets, we demonstrate that RippleNet achieves substantial gains in a variety of scenarios, including movie, book and news recommendation, over several state-of-the-art baselines.

Recent years have witnessed a surge of interest on response generation for neural conversation systems. Most existing models are implemented by following the Encoder-Decoder framework and operate sentences of conversations at word-level. The word-level model is suffering from the Unknown Words Issue and the Preference Issue, which seriously impact the quality of generated responses, for example, generated responses may become irrelevant or too general (i.e. safe responses). To address these issues, this paper proposes a hybrid-level Encoder-Decoder model (HL-EncDec), which not only utilizes the word-level features but also character-level features. We conduct several experiments to evaluate HL-EncDec on a Chinese corpus, experimental results show our model significantly outperforms other non-word-level models in automatic metrics and human annotations and is able to generate more informative responses. We also conduct experiments with a small-scale English dataset to show the generalization ability.

Recently, matrix factorization-based recommendation methods have been criticized for the problem raised by the triangle inequality violation. Although several metric learning-based approaches have been proposed to overcome this issue, existing approaches typically project each user to a single point in the metric space, and thus do not suffice for properly modeling the intensity and the heterogeneity of user-item relationships in implicit feedback. In this paper, we propose TransCF to discover such latent user-item relationships embodied in implicit user-item interactions. Inspired by the translation mechanism popularized by knowledge graph embedding, we construct user-item specific translation vectors by employing the neighborhood information of users and items, and translate each user toward items according to the user's relationships with the items. Our proposed method outperforms several state-of-the-art methods for top-N recommendation on seven real-world data by up to 17% in terms of hit ratio. We also conduct extensive qualitative evaluations on the translation vectors learned by our proposed method to ascertain the benefit of adopting the translation mechanism for implicit feedback-based recommendations.

Explainable recommendation, which provides explanations about why an item is recommended, has attracted increasing attention due to its ability in helping users make better decisions and increasing users' trust in the system. Existing explainable recommendation methods either ignore the working mechanism of the recommendation model or are designed for a specific recommendation model. Moreover, it is difficult for existing methods to ensure the presentation quality of the explanations (e.g., consistency). To solve these problems, we design a reinforcement learning framework for explainable recommendation. Our framework can explain any recommendation model (model-agnostic) and can flexibly control the explanation quality based on the application scenario. To demonstrate the effectiveness of our framework, we show how it can be used for generating sentence-level explanations. Specifically, we instantiate the explanation generator in the framework with a personalized-attention-based neural network. Offline experiments demonstrate that our method can well explain both collaborative filtering methods and deep-learning-based models. Evaluation with human subjects shows that the explanations generated by our method are significantly more useful than the explanations generated by the baselines.

Sentence-level sentiment classification aims to mine fine-grained sentiment information from texts. Existing methods for this task are usually based on supervised learning and rely on massive labeled sentences for model training. However, annotating sufficient sentences is expensive and time-consuming. In this paper, we propose a neural sentence-level sentiment classification approach which can exploit heterogeneous sentiment supervision and reduce the dependence on labeled sentences. Besides the sentence-level supervision from labeled sentences, our approach can also incorporate the word-level supervision extracted from sentiment lexicons, document-level supervision extracted from labeled documents and sentiment relations between sentences extracted from unlabeled documents. A unified neural framework is proposed to fuse heterogeneous sentiment supervision to train sentence-level sentiment classification model. Experiments on benchmark datasets validate the effectiveness of our approach.

Millions of news articles emerge every day. How to provide personalized news recommendations has become a critical task for service providers. In the past few decades, latent factor models has been widely used for building recommender systems (RSs). With the remarkable success of deep learning techniques especially in visual computing and natural language understanding, more and more researchers have been trying to leverage deep neural networks to learn latent representations for advanced RSs. Following mainstream deep learning-based RSs, we propose a novel deep fusion model (DFM), which aims to improve the representation learning abilities in deep RSs and can be used for both candidate retrieval and item re-ranking. There are two key components in our DFM approach, namely an inception module and an attention mechanism. The inception module improves the plain multi-layer network via leveraging of various levels of interaction simultaneously, while the attention mechanism merges latent representations learnt from different channels in a customized fashion. We conduct extensive experiments on a commercial news reading dataset, and the results demonstrate that the proposed DFM is superior to several state-of-the-art models.

Chronic diseases like cancer and diabetes are major threats to human life. Understanding the distribution and progression of chronic diseases of a population is important in assisting the allocation of medical resources as well as the design of policies in preemptive healthcare. Traditional methods to obtain large scale indicators on population health, e.g., surveys and statistical analysis, can be costly and time-consuming and often lead to a coarse spatio-temporal picture. In this paper, we leverage a dataset describing the human mobility patterns of citizens in a large metropolitan area. By viewing local human lifestyles we predict the evolution rate of several chronic diseases at the level of a city neighborhood. We apply the combination of a collaborative topic modeling (CTM) and a Gaussian mixture method (GMM) to tackle the data sparsity challenge and achieve robust predictions on health conditions simultaneously. Our method enables the analysis and prediction of disease rate evolution at fine spatio-temporal scales and demonstrates the potential of incorporating datasets from mobile web sources to improve population health monitoring. Evaluations using real-world check-in and chronic disease morbidity datasets in the city of London show that the proposed CTM+GMM model outperforms various baseline methods.

With a large amount of user activity data accumulated, it is crucial to exploit user sequential behavior for sequential recommendations. Conventionally, user general taste and recent demand are combined to promote recommendation performances. However, existing methods often neglect that user long-term preference keep evolving over time, and building a static representation for user general taste may not adequately reflect the dynamic characters. Moreover, they integrate user-item or item-item interactions through a linear way which limits the capability of model. To this end, in this paper, we propose a novel two-layer hierarchical attention network, which takes the above properties into account, to recommend the next item user might be interested. Specifically, the first attention layer learns user long-term preferences based on the historical purchased item representation, while the second one outputs final user representation through coupling user long-term and short-term preferences. The experimental study demonstrates the superiority of our method compared with other state-of-the-art ones.

Information network embedding is an effective way for efficient graph analytics. However, it still faces with computational challenges in problems such as link prediction and node recommendation, particularly with increasing scale of networks. Hashing is a promising approach for accelerating these problems by orders of magnitude. However, no prior studies have been focused on seeking binary codes for information networks to preserve high-order proximity. Since matrix factorization (MF) unifies and outperforms several well-known embedding methods with high-order proximity preserved, we propose a MF-based \underlineI nformation \underlineN etwork \underlineH ashing (INH-MF) algorithm, to learn binary codes which can preserve high-order proximity. We also suggest Hamming subspace learning, which only updates partial binary codes each time, to scale up INH-MF. We finally evaluate INH-MF on four real-world information network datasets with respect to the tasks of node classification and node recommendation. The results demonstrate that INH-MF can perform significantly better than competing learning to hash baselines in both tasks, and surprisingly outperforms network embedding methods, including DeepWalk, LINE and NetMF, in the task of node recommendation. The source code of INH-MF is available online\footnote\urlhttps://github.com/DefuLian/network .

Combinatorial features are essential for the success of many commercial models. Manually crafting these features usually comes with high cost due to the variety, volume and velocity of raw data in web-scale systems. Factorization based models, which measure interactions in terms of vector product, can learn patterns of combinatorial features automatically and generalize to unseen features as well. With the great success of deep neural networks (DNNs) in various fields, recently researchers have proposed several DNN-based factorization model to learn both low- and high-order feature interactions. Despite the powerful ability of learning an arbitrary function from data, plain DNNs generate feature interactions implicitly and at the bit-wise level. In this paper, we propose a novel Compressed Interaction Network (CIN), which aims to generate feature interactions in an explicit fashion and at the vector-wise level. We show that the CIN share some functionalities with convolutional neural networks (CNNs) and recurrent neural networks (RNNs). We further combine a CIN and a classical DNN into one unified model, and named this new model eXtreme Deep Factorization Machine (xDeepFM). On one hand, the xDeepFM is able to learn certain bounded-degree feature interactions explicitly; on the other hand, it can learn arbitrary low- and high-order feature interactions implicitly. We conduct comprehensive experiments on three real-world datasets. Our results demonstrate that xDeepFM outperforms state-of-the-art models. We have released the source code of xDeepFM at https://github.com/Leavingseason/xDeepFM.

Collaborator Recommendation is a useful application in exploiting big academic data. However, existing works leave out the contextual restriction (i.e., research topics) of people's academic collaboration, thus cannot recommend suitable collaborators for the required research topics. In this work, we propose Context-aware Collaborator Recommendation (CACR), which aims to recommend high-potential new collaborators for people's context-restricted requests. To this end, we design a novel recommendation framework, which consists of two fundamental components: the Collaborative Entity Embedding network (CEE) and the Hierarchical Factorization Model (HFM). In particular, CEE jointly represents researchers and research topics as compact vectors based on their co-occurrence relationships, whereby capturing researchers' context-aware collaboration tendencies and topics' underlying semantics. Meanwhile, HFM extracts researchers' activenesses and conservativenesses, which reflect their intensities of making academic collaborations and tendencies of working with non-collaborated fellows. The extracted activenesses and conservativenesses work collaboratively with the context-aware collaboration tendencies, such that high-quality recommendation can be produced. Extensive experimental studies are conducted with large-scale academic data, whose results verify the effectiveness of our proposed approaches.

Transcribing content from structural images, e.g., writing notes from music scores, is a challenging task as not only the content objects should be recognized, but the internal structure should also be preserved. Existing image recognition methods mainly work on images with simple content (e.g., text lines with characters), but are not capable to identify ones with more complex content (e.g., structured code), which often follow a fine-grained grammar. To this end, in this paper, we propose a hierarchical Spotlight Transcribing Network (STN) framework followed by a two-stage "where-to-what'' solution. Specifically, we first decide "where-to-look'' through a novel spotlight mechanism to focus on different areas of the original image following its structure. Then, we decide "what-to-write'' by developing a GRU based network with the spotlight areas for transcribing the content accordingly. Moreover, we propose two implementations on the basis of STN, i.e., STNM and STNR, where the spotlight movement follows the Markov property and Recurrent modeling, respectively. We also design a reinforcement method to refine our STN framework by self-improving the spotlight mechanism. We conduct extensive experiments on many structural image datasets, where the results clearly demonstrate the effectiveness of STN framework.

Latent Factor Models (LFMs) based on Collaborative Filtering (CF) have been widely applied in many recommendation systems, due to their good performance of prediction accuracy. In addition to users' ratings, auxiliary information such as item features is often used to improve performance, especially when ratings are very sparse. To the best of our knowledge, most existing LFMs integrate different item features in the same way for all users. Nevertheless, the attention on different item attributes varies a lot from user to user. For personalized recommendation, it is valuable to know what feature of an item a user cares most about. Besides, the latent vectors used to represent users or items in LFMs have few explicit meanings, which makes it difficult to explain why an item is recommended to a specific user. In this work, we propose the Attention-driven Factor Model (AFM), which can not only integrate item features driven by users' attention but also help answer this "why". To estimate users' attention distributions on different item features, we propose the Gated Attention Units (GAUs) for AFM. The GAUs make it possible to let the latent factors "talk", by generating user attention distributions from user latent vectors. With users' attention distributions, we can tune the weights of item features for different users. Moreover, users' attention distributions can also serve as explanations for our recommendations. Experiments on several real-world datasets demonstrate the advantages of AFM (using GAUs) over competitive baseline algorithms on rating prediction.

Excellent recommendation system facilitates users retrieving contents they like and, what»s much more important - the contents they might like but they are not aware of yet. It will further increase the satisfaction of users and increase the retention rate and conversion rate indirectly. While the public's now listening to all kinds of music, recommendation algorithms still struggle in key areas. Without enough historical data, how would an algorithm know if listeners will like a new song or a new artist? And, how would it know what songs to recommend brand new users? In WSDM Cup 2018, the first task is to solve the abovementioned challenges to build a better music recommendation system. The 2nd task in the Cup focuses on churn prediction. For a subscription business, accurately predicting churn is critical to long-term success. Even slight variations in churn can drastically affect profits. In this task, participants are asked to build an algorithm that predicts whether a user will churn after their subscription expires. The competition data and award are provided by KKBOX, a leading music streaming service in Taiwan.

In online social networks people often express attitudes towards others, which forms massive sentiment links among users. Predicting the sign of sentiment links is a fundamental task in many areas such as personal advertising and public opinion analysis. Previous works mainly focus on textual sentiment classification, however, text information can only disclose the "tip of the iceberg»» about users» true opinions, of which the most are unobserved but implied by other sources of information such as social relation and users» profile. To address this problem, in this paper we investigate how to predict possibly existing sentiment links in the presence of heterogeneous information. First, due to the lack of explicit sentiment links in mainstream social networks, we establish a labeled heterogeneous sentiment dataset which consists of users» sentiment relation, social relation and profile knowledge by entity-level sentiment extraction method. Then we propose a novel and flexible end-to-end Signed Heterogeneous Information Network Embedding (SHINE) framework to extract users» latent representations from heterogeneous networks and predict the sign of unobserved sentiment links. SHINE utilizes multiple deep autoencoders to map each user into a low-dimension feature space while preserving the network structure. We demonstrate the superiority of SHINE over state-of-the-art baselines on link prediction and node recommendation in two real-world datasets. The experimental results also prove the efficacy of SHINE in cold start scenario.

Data sparsity is one of the biggest problems faced by collaborative filtering used in recommender systems. Data imputation alleviates the data sparsity problem by inferring missing ratings and imputing them to the original rating matrix. In this paper, we identify the limitations of existing data imputation approaches and suggest three new claims that all data imputation approaches should follow to achieve high recommendation accuracy. Furthermore, we propose a deep-learning based approach to compute imputed values that satisfies all three claims. Based on our hypothesis that most pre-use preferences (e.g., impressions) on items lead to their post-use preferences (e.g., ratings), our approach tries to understand via deep learning how pre-use preferences lead to post-use preferences differently depending on the characteristics of users and items. Through extensive experiments on real-world datasets, we verify our three claims and hypothesis, and also demonstrate that our approach significantly outperforms existing state-of-the-art approaches.

Online news recommender systems aim to address the information explosion of news and make personalized recommendation for users. In general, news language is highly condensed, full of knowledge entities and common sense. However, existing methods are unaware of such external knowledge and cannot fully discover latent knowledge-level connections among news. The recommended results for a user are consequently limited to simple patterns and cannot be extended reasonably. To solve the above problem, in this paper, we propose a deep knowledge-aware network (DKN) that incorporates knowledge graph representation into news recommendation. DKN is a content-based deep recommendation framework for click-through rate prediction. The key component of DKN is a multi-channel and word-entity-aligned knowledge-aware convolutional neural network (KCNN) that fuses semantic-level and knowledge-level representations of news. KCNN treats words and entities as multiple channels, and explicitly keeps their alignment relationship during convolution. In addition, to address users» diverse interests, we also design an attention module in DKN to dynamically aggregate a user»s history with respect to current candidate news. Through extensive experiments on a real online news platform, we demonstrate that DKN achieves substantial gains over state-of-the-art deep recommendation models. We also validate the efficacy of the usage of knowledge in DKN.

In this paper, we propose a novel Deep Reinforcement Learning framework for news recommendation. Online personalized news recommendation is a highly challenging problem due to the dynamic nature of news features and user preferences. Although some online recommendation models have been proposed to address the dynamic nature of news recommendation, these methods have three major issues. First, they only try to model current reward (e.g., Click Through Rate). Second, very few studies consider to use user feedback other than click / no click labels (e.g., how frequent user returns) to help improve recommendation. Third, these methods tend to keep recommending similar news to users, which may cause users to get bored. Therefore, to address the aforementioned challenges, we propose a Deep Q-Learning based recommendation framework, which can model future reward explicitly. We further consider user return pattern as a supplement to click / no click label in order to capture more user feedback information. In addition, an effective exploration strategy is incorporated to find new attractive news for users. Extensive experiments are conducted on the offline dataset and online production environment of a commercial news recommendation application and have shown the superior performance of our methods.

Location recommendation plays an essential role in helping people find attractive places. Though recent research has studied how to recommend locations with social and geographical information, few of them addressed the cold-start problem of new users. Because mobility records are often shared on social networks, semantic information can be leveraged to tackle this challenge. A typical method is to feed them into explicit-feedback-based content-aware collaborative filtering, but they require drawing negative samples for better learning performance, as users’ negative preference is not observable in human mobility. However, prior studies have empirically shown sampling-based methods do not perform well. To this end, we propose a scalable Implicit-feedback-based Content-aware Collaborative Filtering (ICCF) framework to incorporate semantic content and to steer clear of negative sampling. We then develop an efficient optimization algorithm, scaling linearly with data size and feature size, and quadratically with the dimension of latent space. We further establish its relationship with graph Laplacian regularized matrix factorization. Finally, we evaluate ICCF with a large-scale LBSN dataset in which users have profiles and textual content. The results show that ICCF outperforms several competing baselines, and that user information is not only effective for improving recommendations but also coping with cold-start scenarios.

As the worlds of commerce and the Internet technology become more inextricably linked, a large number of user consumption series become available for online market intelligence analysis. A critical demand along this line is to predict the future product adoption state of each user, which enables a wide range of applications such as targeted marketing. Nevertheless, previous works only aimed at predicting if a user would adopt a particular product or not with a binary buy-or-not representation. The problem of tracking and predicting users' adoption rates, i.e., the frequency and regularity of using each product over time, is still under-explored. To this end, we present a comprehensive study of product adoption rate prediction in a competitive market. This task is nontrivial as there are three major challenges in modeling users' complex adoption states: the heterogeneous data sources around users, the unique user preference and the competitive product selection. To deal with these challenges, we first introduce a flexible factor-based decision function to capture the change of users' product adoption rate over time, where various factors that may influence users' decisions from heterogeneous data sources can be leveraged. Using this factor-based decision function, we then provide two corresponding models to learn the parameters of the decision function with both generalized and personalized assumptions of users' preferences. We further study how to leverage the competition among different products and simultaneously learn product competition and users' preferences with both generalized and personalized assumptions. Finally, extensive experiments on two real-world datasets show the superiority of our proposed models.

Location recommendation is an important means to help people discover attractive locations. However, extreme sparsity of user-location matrices leads to a severe challenge, so it is necessary to take implicit feedback characteristics of user mobility data into account and leverage the location’s spatial information. To this end, based on previously developed GeoMF, we propose a scalable and flexible framework, dubbed GeoMF++, for joint geographical modeling and implicit feedback-based matrix factorization. We then develop an efficient optimization algorithm for parameter learning, which scales linearly with data size and the total number of neighbor grids of all locations. GeoMF++ can be well explained from two perspectives. First, it subsumes two-dimensional kernel density estimation so that it captures spatial clustering phenomenon in user mobility data; Second, it is strongly connected with widely used neighbor additive models, graph Laplacian regularized models, and collective matrix factorization. Finally, we extensively evaluate GeoMF++ on two large-scale LBSN datasets. The experimental results show that GeoMF++ consistently outperforms the state-of-the-art and other competing baselines on both datasets in terms of NDCG and Recall. Besides, the efficiency studies show that GeoMF++ is much more scalable with the increase of data size and the dimension of latent space.

Online Judge (OJ) systems have been widely used in many areas, including programming, mathematical problems solving, and job interviews. Unlike other online learning systems, such as Massive Open Online Course, most OJ systems are designed for self-directed learning without the intervention of teachers. Also, in most OJ systems, problems are simply listed in volumes and there is no clear organization of them by topics or difficulty levels. As such, problems in the same volume are mixed in terms of topics or difficulty levels. By analyzing large-scale users’ learning traces, we observe that there are two major learning modes (or patterns). Users either practice problems in a sequential manner from the same volume regardless of their topics or they attempt problems about the same topic, which may spread across multiple volumes. Our observation is consistent with the findings in classic educational psychology. Based on our observation, we propose a novel two-mode Markov topic model to automatically detect the topics of online problems by jointly characterizing the two learning modes. For further predicting the difficulty level of online problems, we propose a competition-based expertise model using the learned topic information. Extensive experiments on three large OJ datasets have demonstrated the effectiveness of our approach in three different tasks, including skill topic extraction, expertise competition prediction and problem recommendation.

The analysis and study of the relationship between a geo-spatial event and human mobility in an urban area is very significant for improving productivity, mobility, and safety. In particular, in order to alleviate serious road congestions, traffic jams, and stampedes, it is essential to predict and be informed about the occurrence of an event as soon as possible. When we know an event occurrence in advance, some of those who are not interested in the event might change their plans and/or might take a detour to avoid to get involved in a heavy congestion. In this context, this paper presents an early event detection technique using GPS trajectories collected from periodic-cars, which are vehicles periodically traveling on a pre-scheduled route with a pre-determined departure time, such as a transit bus, shuttle, garbage truck, or municipal patrol car. Using these trajectories, which provide the real-time and continuous traffic flow and speed, our technique detects large-scale events in advance, without incurring any privacy invasion. The behavior of periodic-cars shows a certain sign of a large-scale event before attendees gather around a venue because traffic can be slowed around the venue before the event occurrence. We evaluated our method using over 7,000-bus data from January to May in 2015 in Beijing, which we compared with the check-in data collected from a social network service.

Product sales forecasting enables comprehensive understanding of products' future development, making it of particular interest for companies to improve their business, for investors to measure the values of firms, and for users to capture the trends of a market. Recent studies show that the complex competition interactions among products directly influence products' future development. However, most existing approaches fail to model the evolutionary competition among products and lack the capability to organically reflect multi-level competition analysis in sales forecasting. To address these problems, we propose the Evolutionary Hierarchical Competition Model (EHCM), which effectively considers the time-evolving multi-level competition among products. The EHCM model systematically integrates hierarchical competition analysis with multi-scale time series forecasting. Extensive experiments using a real-world app download dataset show that EHCM outperforms state-of-the-art methods in various forecasting granularities.

Technologies are increasingly taking advantage of the explosion in the amount of data generated by social multimedia (e.g., web searches, ad targeting, and urban computing). In this paper, we propose a multi-view learning framework for presenting the construction of a new urban movement knowledge graph, which could greatly facilitate the research domains mentioned above. In particular, by viewing GPS trajectory data from temporal, spatial, and spatiotemporal points of view, we construct a knowledge graph of which nodes and edges are their locations and relations, respectively. On the knowledge graph, both nodes and edges are represented in latent semantic space. We verify its utility by subsequently applying the knowledge graph to predict the extent of user attention (high or low) paid to different locations in a city. Experimental evaluations and analysis of a real-world dataset show significant improvements in comparison to state-of-the-art methods.

Precisely recommending relevant items from massive candidates to a large number of users is an indispensable yet computationally expensive task in many online platforms (e.g., Amazon.com and Netflix.com). A promising way is to project users and items into a Hamming space and then recommend items via Hamming distance. However, previous studies didn't address the cold-start challenges and couldn't make the best use of preference data like implicit feedback. To fill this gap, we propose a Discrete Content-aware Matrix Factorization (DCMF) model, 1) to derive compact yet informative binary codes at the presence of user/item content information; 2) to support the classification task based on a local upper bound of logit loss; 3) to introduce an interaction regularization for dealing with the sparsity issue. We further develop an efficient discrete optimization algorithm for parameter learning. Based on extensive experiments on three real-world datasets, we show that DCFM outperforms the state-of-the-arts on both regression and classification tasks.

An incisive understanding of user personality is not only essential to many scientific disciplines, but also has a profound business impact on practical applications such as digital marketing, personalized recommendation, mental diagnosis, and human resources management. Previous studies have demonstrated that language usage in social media is effective in personality prediction. However, except for single language features, a less researched direction is how to leverage the heterogeneous information on social media to have a better understanding of user personality. In this paper, we propose a Heterogeneous Information Ensemble framework, called HIE, to predict users' personality traits by integrating heterogeneous information including self-language usage, avatar, emoticon, and responsive patterns. In our framework, to improve the performance of personality prediction, we have designed different strategies extracting semantic representations to fully leverage heterogeneous information on social media. We evaluate our methods with extensive experiments based on a real-world data covering both personality survey results and social media usage from thousands of volunteers. The results reveal that our approaches significantly outperform several widely adopted state-of-the-art baseline methods. To figure out the utility of HIE in a real-world interactive setting, we also present DiPsy, a personalized chatbot to predict user personality through heterogeneous information in digital traces and conversation logs.

Last decades have witnessed a vast amount of interest and research in recommendation systems. Collaborative filtering, which uses the known preferences of a group of users to make recommendations or predictions of the unknown preferences for other users, is one of the most successful approaches to build recommendation systems. Most previous collaborative filtering approaches employ the matrix factorization techniques to learn latent user feature profiles and item feature profiles. Also many subsequent works are proposed to incorporate users' social network information and items' attributions to further improve recommendation performance under the matrix factorization framework. However, the matrix factorization based methods may not make full use of the rating information, leading to unsatisfying performance. Recently deep learning has been approved to be able to find good representations in natural language processing, image classification, and so on. Along this line, we propose a collaborative ranking framework via representation learning with pair-wise constraints (REAP for short), in which autoencoder is used to simultaneously learn the latent factors of both users and items and pair-wise ranked loss defined by (user, item) pairs is considered. Extensive experiments are conducted on five data sets to demonstrate the effectiveness of the proposed framework.

Microblogging Web sites, such as Twitter and Sina Weibo, have become popular platforms for socializing and sharing information in recent years. Spammers have also discovered this new opportunity to unfairly overpower normal users with unsolicited content, namely social spams. Although it is intuitive for everyone to follow legitimate users, recent studies show that both legitimate users and spammers follow spammers for different reasons. Evidence of users seeking spammers on purpose is also observed. We regard this behavior as useful information for spammer detection. In this article, we approach the problem of spammer detection by leveraging the “carefulness” of users, which indicates how careful a user is when she is about to follow a potential spammer. We propose a framework to measure the carefulness and develop a supervised learning algorithm to estimate it based on known spammers and legitimate users. We illustrate how the robustness of the detection algorithms can be improved with aid of the proposed measure. Evaluation on two real datasets from Sina Weibo and Twitter with millions of users are performed, as well as an online test on Sina Weibo. The results show that our approach indeed captures the carefulness, and it is effective for detecting spammers. In addition, we find that our measure is also beneficial for other applications, such as link prediction.

editorial Free Access Share on Mobile Social Multimedia Analytics in the Big Data Era: An Introduction to the Special Issue Editors: Rongrong Ji Xiamen University, China Xiamen University, ChinaView Profile , Wei Liu Tencent AI Lab, China Tencent AI Lab, ChinaView Profile , Xing Xie Microsoft Research Asia, China Microsoft Research Asia, ChinaView Profile , Yiqiang Chen Chinese Academy of Science, China Chinese Academy of Science, ChinaView Profile , Jiebo Luo University of Rochester, United States University of Rochester, United StatesView Profile Authors Info & Claims ACM Transactions on Intelligent Systems and TechnologyVolume 8Issue 3May 2017 Article No.: 34pp 1–3https://doi.org/10.1145/3040934Published:14 April 2017Publication History 2citation224DownloadsMetricsTotal Citations2Total Downloads224Last 12 Months11Last 6 weeks2 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF

In recent decades, the frequency and intensity of natural disasters has increased significantly, and this trend is expected to continue. Therefore, understanding and predicting human behavior and mobility during a disaster will play a vital role in planning effective humanitarian relief, disaster management, and long-term societal reconstruction. However, such research is very difficult to perform owing to the uniqueness of various disasters and the unavailability of reliable and large-scale human mobility data. In this study, we collect big and heterogeneous data (e.g., GPS records of 1.6 million users 1 over 3 years, data on earthquakes that have occurred in Japan over 4 years, news report data, and transportation network data) to study human mobility following natural disasters. An empirical analysis is conducted to explore the basic laws governing human mobility following disasters, and an effective human mobility model is developed to predict and simulate population movements. The experimental results demonstrate the efficiency of our model, and they suggest that human mobility following disasters can be significantly more predictable and be more easily simulated than previously thought.

The identification of network motifs has important applications in numerous domains, such as pattern detection in biological networks and graph analysis in digital circuits. However, mining network motifs is computationally challenging, as it requires enumerating subgraphs from a real-life graph, and computing the frequency of each subgraph in a large number of random graphs. In particular, existing solutions often require days to derive network motifs from biological networks with only a few thousand vertices. To address this problem, this paper presents a novel study on network motif discovery using Graphical Processing Units (GPUs). The basic idea is to employ GPUs to parallelize a large number of subgraph matching tasks in computing subgraph frequencies from random graphs, so as to reduce the overall computation time of network motif discovery. We explore the design space of GPU-based subgraph matching algorithms, with careful analysis of several crucial factors (such as branch divergences and memory coalescing) that affect the performance of GPU programs. Based on our analysis, we develop a GPU-based solution that (i) considerably differs from existing CPU-based methods in how it enumerates subgraphs, and (ii) exploits the strengths of GPUs in terms of parallelism while mitigating their limitations in terms of the computation power per GPU core. With extensive experiments on a variety of biological networks, we show that our solution is up to two orders of magnitude faster than the best CPU-based approach, and is around 20 times more cost-effective than the latter, when taking into account the monetary costs of the CPU and GPUs used.

The frequency and intensity of natural disasters has increased significantly in recent decades, and this trend is expected to continue. Hence, understanding and predicting human evacuation behavior and mobility will play a vital role in planning effective humanitarian relief, disaster management, and long-term societal reconstruction. However, existing models are shallow models, and it is difficult to apply them for understanding the “deep knowledge” of human mobility. Therefore, in this study, we collect big and heterogeneous data (e.g., GPS records of 1.6 million users over 3 years, data on earthquakes that have occurred in Japan over 4 years, news report data, and transportation network data), and we build an intelligent system, namely, DeepMob, for understanding and predicting human evacuation behavior and mobility following different types of natural disasters. The key component of DeepMob is based on a deep learning architecture that aims to understand the basic laws that govern human behavior and mobility following natural disasters, from big and heterogeneous data. Furthermore, based on the deep learning model, DeepMob can accurately predict or simulate a person’s future evacuation behaviors or evacuation routes under different disaster conditions. Experimental results and validations demonstrate the efficiency and superior performance of our system, and suggest that human mobility following disasters may be predicted and simulated more easily than previously thought.

Intelligent personal assistants on mobile devices such as Apple’s Siri and Microsoft Cortana are increasingly important. Instead of passively reacting to queries, they provide users with brand new proactive experiences that aim to offer the right information at the right time. It is, therefore, crucial for personal assistants to understand users’ intent, that is, what information users need now. Intent is closely related to context. Various contextual signals, including spatio-temporal information and users’ activities, can signify users’ intent. It is, however, challenging to model the correlation between intent and context. Intent and context are highly dynamic and often sequentially correlated. Contextual signals are usually sparse, heterogeneous, and not simultaneously available. We propose an innovative collaborative nowcasting model to jointly address all these issues. The model effectively addresses the complex sequential and concurring correlation between context and intent and recognizes users’ real-time intent with continuously arrived contextual signals. We extensively evaluate the proposed model with real-world data sets from a commercial personal assistant. The results validate the effectiveness the proposed model, and demonstrate its capability of handling the real-time flow of contextual signals. The studied problem and model also provide inspiring implications for new paradigms of recommendation on mobile intelligent devices.

To address the difficulty in clipping articles from various mobile applications (apps), we propose a novel framework called UniClip, which allows a user to snap a screen of an article to save the whole article in one place. The key task of the framework is search by screenshots , which has three challenges: (1) how to represent a screenshot; (2) how to formulate queries for effective article retrieval; and (3) how to identify the article from search results. We solve these by (1) segmenting a screenshot into structural units called blocks, (2) formulating effective search queries by considering the role of each block, and (3) aggregating the search result lists of multiple queries. To improve efficiency, we also extend our approach with learning-to-rank techniques so that we can find the desired article with only one query. Experimental results show that our approach achieves high retrieval performance ( F 1 = 0.868), which outperforms baselines based on keyword extraction and chunking methods. Learning-to-rank models improve our approach without learning by about 6%. A user study conducted to investigate the usability of UniClip reveals that ours is preferred by 21 out of 22 participants for its simplicity and effectiveness.

People often use multiple platforms to fulfill their different information needs. With the ultimate goal of serving people intelligently, a fundamental way is to get comprehensive understanding about user needs. How to organically integrate and bridge cross-platform information in a human-centric way is important. Existing transfer learning assumes either fully-overlapped or non-overlapped among the users. However, the real case is the users of different platforms are partially overlapped. The number of overlapped users is often small and the explicitly known overlapped users is even less due to the lacking of unified ID for a user across different platforms. In this paper, we propose a novel semi-supervised transfer learning method to address the problem of cross-platform behavior prediction, called XPTrans. To alleviate the sparsity issue, it fully exploits the small number of overlapped crowds to optimally bridge a user's behaviors in different platforms. Extensive experiments across two real social networks show that XPTrans significantly outperforms the state-of-the-art. We demonstrate that by fully exploiting 26% overlapped users, XPTrans can predict the behaviors of non-overlapped users with the same accuracy as overlapped users, which means the small overlapped crowds can successfully bridge the information across different platforms.

What people buy is an important aspect or view of lifestyles. Studying people's shopping patterns in different urban regions can not only provide valuable information for various commercial opportunities, but also enable a better understanding about urban infrastructure and urban lifestyle. In this paper, we aim to predict citywide shopping patterns. This is a challenging task due to the sparsity of the available data -- over 60% of the city regions are unknown for their shopping records. To address this problem, we incorporate another important view of human lifestyles, namely mobility patterns. With information on "where people go", we infer "what people buy". Moreover, to model the relations between regions, we exploit spatial interactions in our method. To that end, Collective Matrix Factorization (CMF) with an interaction regularization model is applied to fuse the data from multiple views or sources. Our experimental results have shown that our model outperforms the baseline methods on two standard metrics. Our prediction results on multiple shopping patterns reveal the divergent demands in different urban regions, and thus reflect key functional characteristics of a city. Furthermore, we are able to extract the connection between the two views of lifestyles, and achieve a better or novel understanding of urban lifestyles.

Fine-grained long-term forecasting enables many emerging recommendation applications such as forecasting the usage amounts of various apps to guide future investments, and forecasting users' seasonal demands for a certain commodity to find potential repeat buyers. For these applications, there often exists certain homogeneity in terms of similar users and items (e.g., apps), which also correlates with various contexts like users' spatial movements and physical environments. Most existing works only focus on predicting the upcoming situation such as the next used app or next online purchase, without considering the long-term temporal co-evolution of items and contexts and the homogeneity among all dimensions. In this paper, we propose a contextual collaborative forecasting (CCF) model to address the above issues. The model integrates contextual collaborative filtering with time series analysis, and simultaneously captures various components of temporal patterns, including trend, seasonality, and stationarity. The approach models the temporal homogeneity of similar users, items, and contexts. We evaluate the model on a large real-world app usage dataset, which validates that CCF outperforms state-of-the-art methods in terms of both accuracy and efficiency for long-term app usage forecasting.

The prediction of academic performance is one of the most important tasks in educational data mining, and has been widely studied in MOOCs and intelligent tutoring systems. Academic performance could be affected with factors like personality, skills, social environment, the use of library books and so on. However, it is still less investigated that how could the use of library books affect academic performance of college students and even leverage book-loan history for predicting academic performance. To this end, we propose a supervised content-aware matrix factorization for mutual reinforcement of academic performance prediction and library book recommendation. This model not only addresses the sparsity challenge by explainable dimension reduction techniques, but also promotes library book recommendation by recommending "right" books for students based on their performance levels and book meta information. Finally, we evaluate the proposed model on three years of the book-loan history and cumulative grade point average of 13,047 undergraduate students in one university. The results show that the proposed model outperforms the competing baselines on both tasks, and that academic performance is not only predictable from the book-loan history but also improves the recommendation of library books for students.

Although weighted tensor factorization tailored to implicit feedback has shown its superior performance in temporal-aware location recommendation, it suffers from three critical challenges. First, it doesn't distinguish the confidence of negative preference for time-dependent unvisited locations from that for fully unvisited ones. Second, discontinuity arises from time discretization, and thus an infinitely large margin may exist between different bins of time. Third, geographical constraints of neighbor locations are not taken into account. To address these challenges, we propose a regularized content-aware tensor factorization (RCTF) algorithm, which exploits three strategies to address the corresponding challenges. First, it introduces a novel interaction regularization, second, it represents each bin of time by a derived feature vector from eigen decomposition of a time-bin similarity matrix, to capture the proximity of neighbor bins of time, third, it encodes geographical information of locations by discrete spatial distributions, so that spatial proximity constraints can be satisfied by simply feeding them into location content. The proposed algorithm is then evaluated for time-aware location recommendation on two large scale location-based social network datasets. The experimental results show the superiority of the proposed algorithm to several competing time-aware recommendation baselines, and verify the significant benefit of three strategies in the proposed algorithm.

Matrix completion is the task of recovering a data matrix from a sample of entries, and has received significant attention in theory and practice. Normally, matrix completion considers a single matrix, which can be a noisy image or a rating matrix in recommendation. In practice however, data is often obtained from multiple domains rather than a single domain. For example, in recommendation, multiple matrices may exist as user x movie and user x book, while correlations among the multiple domains can be reasonably exploited to improve the quality of matrix completion. In this paper, we consider the problem of aligned matrix completion, where multiple matrices are recovered that correspond to different representations of the same group of objects. In the proposed model, we maintain consistency of multiple domains with a shared latent structure, while allowing independent patterns for each separate domain. In addition, we impose the low-rank structure of a matrix with a novel regularizer which provides better approximation than the standard nuclear norm relaxation.

A new paradigm of recommendation is emerging in intelligent personal assistants such as Apple's Siri, Google Now, and Microsoft Cortana, which recommends "the right information at the right time" and proactively helps you "get things done". This type of recommendation requires precisely tracking users' contemporaneous intent, i.e., what type of information (e.g., weather, stock prices) users currently intend to know, and what tasks (e.g., playing music, getting taxis) they intend to do. Users' intent is closely related to context, which includes both external environments such as time and location, and users' internal activities that can be sensed by personal assistants. The relationship between context and intent exhibits complicated co-occurring and sequential correlation, and contextual signals are also heterogeneous and sparse, which makes modeling the context intent relationship a challenging task. To solve the intent tracking problem, we propose the Kalman filter regularized PARAFAC2 (KP2) nowcasting model, which compactly represents the structure and co-movement of context and intent. The KP2 model utilizes collaborative capabilities among users, and learns for each user a personalized dynamic system that enables efficient nowcasting of users' intent. Extensive experiments using real-world data sets from a commercial personal assistant show that the KP2 model significantly outperforms various methods, and provides inspiring implications for deploying large-scale proactive recommendation systems in personal assistants.

Among different recommendation techniques, collaborative filtering usually suffer from limited performance due to the sparsity of user-item interactions. To address the issues, auxiliary information is usually used to boost the performance. Due to the rapid collection of information on the web, the knowledge base provides heterogeneous information including both structured and unstructured data with different semantics, which can be consumed by various applications. In this paper, we investigate how to leverage the heterogeneous information in a knowledge base to improve the quality of recommender systems. First, by exploiting the knowledge base, we design three components to extract items' semantic representations from structural content, textual content and visual content, respectively. To be specific, we adopt a heterogeneous network embedding method, termed as TransR, to extract items' structural representations by considering the heterogeneity of both nodes and relationships. We apply stacked denoising auto-encoders and stacked convolutional auto-encoders, which are two types of deep learning based embedding techniques, to extract items' textual representations and visual representations, respectively. Finally, we propose our final integrated framework, which is termed as Collaborative Knowledge Base Embedding (CKE), to jointly learn the latent representations in collaborative filtering as well as items' semantic representations from the knowledge base. To evaluate the performance of each embedding component as well as the whole system, we conduct extensive experiments with two real-world datasets from different scenarios. The results reveal that our approaches outperform several widely adopted state-of-the-art recommendation methods.

Given a set D of tuples defined on a domain Omega, we study differentially private algorithms for constructing a histogram over Omega to approximate the tuple distribution in D. Existing solutions for the problem mostly adopt a hierarchical decomposition approach, which recursively splits Omega into sub-domains and computes a noisy tuple count for each sub-domain, until all noisy counts are below a certain threshold. This approach, however, requires that we (i) impose a limit h on the recursion depth in the splitting of Omega and (ii) set the noise in each count to be proportional to h. The choice of h is a serious dilemma: a small h makes the resulting histogram too coarse-grained, while a large h leads to excessive noise in the tuple counts used in deciding whether sub-domains should be split. Furthermore, h cannot be directly tuned based on D; otherwise, the choice of h itself reveals private information and violates differential privacy. To remedy the deficiency of existing solutions, we present PrivTree, a histogram construction algorithm that adopts hierarchical decomposition but completely eliminates the dependency on a pre-defined h. The core of PrivTree is a novel mechanism that (i) exploits a new analysis on the Laplace distribution and (ii) enables us to use only a constant amount of noise in deciding whether a sub-domain should be split, without worrying about the recursion depth of splitting. We demonstrate the application of PrivTree in modelling spatial data, and show that it can be extended to handle sequence data (where the decision in sub-domain splitting is not based on tuple counts but a more sophisticated measure). Our experiments on a variety of real datasets show that PrivTree considerably outperforms the states of the art in terms of data utility.

Friendships are dynamic. Previous studies have converged to suggest that social interactions, in both online and offline social networks, are diagnostic reflections of friendship relations (also called social ties). However, most existing approaches consider a social tie as either a binary relation, or a fixed value (named tie strength). In this paper, we investigate the dynamics of dyadic friend relationships through online social interactions, in terms of a variety of aspects, such as reciprocity, temporality, and contextuality. In turn, we propose a model to predict repliers and retweeters given a particular tweet posted at a certain time in a microblog-based social network. More specifically, we have devised a learning-to-rank approach to train a ranker that considers elaborate user-level and tweet-level features (like sentiment, self-disclosure, and responsiveness) to address these dynamics. In the prediction phase, a tweet posted by a user is deemed a query and the predicted repliers/retweeters are retrieved using the learned ranker. We have collected a large dataset containing 73.3 million dyadic relationships with their interactions (replies and retweets). Extensive experimental results based on this dataset show that by incorporating the dynamics of friendship relations, our approach significantly outperforms state-of-the-art models in terms of multiple evaluation metrics, such as MAP, NDCG and Topmost Accuracy. In particular, the advantage of our model is even more promising in predicting the exact sequence of repliers/retweeters considering their orders. Furthermore, the proposed approach provides emerging implications for many high-value applications in online social networks.

Mobile digital assistants such as Microsoft Cortana and Google Now currently offer appealing proactive experiences to users, which aim to deliver the right information at the right time. To achieve this goal, it is crucial to precisely predict users' real-time intent. Intent is closely related to context, which includes not only the spatial-temporal information but also users' current activities that can be sensed by mobile devices. The relationship between intent and context is highly dynamic and exhibits chaotic sequential correlation. The context itself is often sparse and heterogeneous. The dynamics and co-movement among contextual signals are also elusive and complicated. Traditional recommendation models cannot directly apply to proactive experiences because they fail to tackle the above challenges. Inspired by the nowcasting practice in meteorology and macroeconomics, we propose an innovative collaborative nowcasting model to effectively resolve these challenges. The proposed model successfully addresses sparsity and heterogeneity of contextual signals. It also effectively models the convoluted correlation within contextual signals and between context and intent. Specifically, the model first extracts collaborative latent factors, which summarize shared temporal structural patterns in contextual signals, and then exploits the collaborative Kalman Filter to generate serially correlated personalized latent factors, which are utilized to monitor each user's real-time intent. Extensive experiments with real-world data sets from a commercial digital assistant demonstrate the effectiveness of the collaborative nowcasting model. The studied problem and model provide inspiring implications for new paradigms of recommendations on mobile intelligent devices.

The wide adoption of location-based services provide the potential to understand people's mobility pattern at an unprecedented level, which can also enable food-service industry to accurately predict consumers' dining behavior. In this paper, based on users' dining implicit feedbacks (restaurant visit via check-ins), explicit feedbacks (restaurant reviews) as well as some meta data (e.g., location, user demographics, restaurant attributes), we aim at recommending each user a list of restaurants for his next dining. Implicit and Explicit feedbacks of dining behavior exhibit different characteristics of user preference. Therefore, in our work, user's dining preference mainly contains two parts: implicit preference coming from check-in data (implicit feedbacks) and explicit preference coming from rating and review data (explicit feedbacks). For implicit preference, we first apply a probabilistic tensor factorization model (PTF) to capture preference in a latent subspace. Then, in order to incorporate contextual signals from meta data, we extend PTF by proposing an Implicit Preference Model (IPM), which can simultaneously capture users'/restaurants'/time' preference in the collaborative filtering and dining preference in a specific context (e.g., spatial distance preference, environmental preference). For explicit preference, we propose Explicit Preference Model (EPM) by combining matrix factorization with topic modeling to discover the user preference embedded both in rating score and text content. Finally, we design a unified model termed as Collective Implicit Explicit Preference Model (CIEPM) to combine implicit and explicit preference together for restaurant recommendation. To evaluate the performance of our system, we conduct extensive experiments with large-scale datasets covering hundreds of thousands of users and restaurants. The results reveal that our system is effective for restaurant recommendation.

In this paper we address the difficulty of clipping articles from mobile apps. We propose a service called UniClip that allows a user to save the full content of an article by snapping a screenshot part of it. UniClip leverages a huge amount of indexed web data to mine the article by starting with a snapped screenshot. We propose approaches to solve three challenges: (1) how to represent a screenshot; (2) how to formulate effective queries for retrieving a full article; and (3) how to rank the best URL at the top from multiple search result lists. Experimental results indicate that our approach is effective in achieving as high an $$F_1$$ F 1 measure as 0.905, which outperforms the best of three baseline methods by 18 points.

Collaborative filtering (CF) models offer users personalized recommendations by measuring the relevance between the active user and each individual candidate item. Following this idea, user-based collaborative filtering (UCF) usually selects the local popular items from the like-minded neighbor users. However, these traditional relevance-based models only consider the individuals (i.e., each neighbor user and candidate item) separately during neighbor set selection and recommendation set generation, thus usually incurring highly similar recommendations that lack diversity. While many researchers have recognized the importance of diversified recommendations, the proposed solutions either needed additional semantic information of items or decreased accuracy in this process. In this article, we describe how to generate both accurate and diversified recommendations from a new perspective. Along this line, we first introduce a simple measure of coverage that quantifies the usefulness of the whole set, that is, the neighbor userset and the recommended itemset as a complete entity. Then we propose a recommendation framework named REC that considers both traditional relevance-based scores and the new coverage measure based on UCF. Under REC, we further prove that the goals of maximizing relevance and coverage measures simultaneously in both the neighbor set selection step and the recommendation set generation step are NP-hard. Luckily, we can solve them effectively and efficiently by exploiting the inherent submodular property. Furthermore, we generalize the coverage notion and the REC framework from both a data perspective and an algorithm perspective. Finally, extensive experimental results on three real-world datasets show that the REC-based recommendation models can naturally generate more diversified recommendations without decreasing accuracy compared to some state-of-the-art models.

With the boom of social media, it is a very popular trend for people to share what they are doing with friends across various social networking platforms. Nowadays, we have a vast amount of descriptions, comments, and ratings for local services. The information is valuable for new users to judge whether the services meet their requirements before partaking. In this paper, we propose a user-service rating prediction approach by exploring social users' rating behaviors. In order to predict user-service ratings, we focus on users' rating behaviors. In our opinion, the rating behavior in recommender system could be embodied in these aspects: 1) when user rated the item, 2) what the rating is, 3) what the item is, 4) what the user interest that we could dig from his/her rating records is, and 5) how the user's rating behavior diffuses among his/her social friends. Therefore, we propose a concept of the rating schedule to represent users' daily rating behaviors. In addition, we propose the factor of interpersonal rating behavior diffusion to deep understand users' rating behaviors. In the proposed user-service rating prediction approach, we fuse four factors-user personal interest (related to user and the item's topics), interpersonal interest similarity (related to user interest), interpersonal rating behavior similarity (related to users' rating behavior habits), and interpersonal rating behavior diffusion (related to users' behavior diffusions)-into a unified matrix-factorized framework. We conduct a series of experiments in the Yelp dataset and Douban Movie dataset. Experimental results show the effectiveness of our approach.

News recommendation has become a big attraction with which major Web search portals retain their users. Two effective approaches are Content-based Filtering and Collaborative Filtering, each serving a specific recommendation scenario. The Content-based Filtering approaches inspect rich contexts of the recommended items, while the Collaborative Filtering approaches predict the interests of long-tail users by collaboratively learning from interests of related users. We have observed empirically that, for the problem of news topic displaying, both the rich context of news topics and the long-tail users exist. Therefore, in this paper, we propose a Content-based Collaborative Filtering approach (CCF) to bring both Content-based Filtering and Collaborative Filtering approaches together. We found that combining the two is not an easy task, but the benefits of CCF are impressive. On one hand, CCF makes recommendations based on the rich contexts of the news. On the other hand, CCF collaboratively analyzes the scarce feedbacks from the long-tail users. We tailored this CCF approach for the news topic displaying on the Bing front page and demonstrated great gains in attracting users. In the experiments and analyses part of this paper, we discuss the performance gains and insights in news topic recommendation in Bing.

The frequency and intensity of natural disasters has significantly increased over the past decades and this trend is predicted to continue. Facing these possible and unexpected disasters, understanding and simulating of human emergency mobility following disasters will becomethe critical issue for planning effective humanitarian relief, disaster management, and long-term societal reconstruction. However, due to the uniquenessof various disasters and the unavailability of reliable and large scale human mobility data, such kind of research is very difficult to be performed. Hence, in this paper,we collect big and heterogeneous data (e.g. 1.6 million users' GPS records in three years, 17520 times of Japan earthquake data in four years, news reporting data, transportation network data and etc.) to capture and analyze human emergency mobility following different disasters. By mining these big data, we aim to understand what basic laws govern human mobility following disasters, and develop a general model of human emergency mobility for generating and simulating large amount of human emergency movements. The experimental results and validations demonstrate the efficiency of our simulation model, and suggest that human mobility following disasters may be significantly morepredictable and can be easier simulated than previously thought.

Consumer impulsivity is a psychological feature characterizing the impulsive buying tendency. In this paper, by bridging consumer behavior with perceived stimuli on social networks, we present a computational framework, termed Consumer Impulsivity Model (CIM), for exploring a consumer's impulsivity in both offline and online context: consumption-related location visit indicating consumption patterns in the physical realm, and online shopping behavior indicating economic activities on the Internet. To demonstrate the effectiveness of CIM, we conduct extensive experiments, with a large dataset we have collected from thousands of consumers. The results show that 1) for 103 subjects, the inferred consumer impulsivity has a positive Pearson correlation with survey results in the situation of product and product category, respectively. 2) females inferred impulsivity is higher than males on average in the situation of product and product category, respectively. Age has a negative Pearson correlation with inferred impulsivity in the situation of POI, POI category and product category, respectively. 3) for next behavior prediction, our model defeats several presented baselines. These results suggest that our framework CIM offers a powerful paradigm for 1) presenting an effective measurement for consumer impulsivity. 2) uncovering the correlation between consumer impulsivity and demographic factors and 3) revealing that the introduction of impulsivity is effective in predicting consumer behavior.

The identification of network motifs has important applications in numerous domains, such as pattern detection in biological networks and graph analysis in digital circuits. However, mining network motifs is computationally challenging, as it requires enumerating subgraphs from a real-life graph, and computing the frequency of each subgraph in a large number of random graphs. In particular, existing solutions often require days to derive network motifs from biological networks with only a few thousand vertices. To address this problem, this paper presents a novel study on network motif discovery using Graphical Processing Units (GPUs). The basic idea is to employ GPUs to parallelize a large number of subgraph matching tasks in computing subgraph frequencies from random graphs, so as to reduce the overall computation time of network motif discovery. We explore the design space of GPU-based subgraph matching algorithms, with careful analysis of several crucial factors that affect the performance of GPU programs. Based on our analysis, we develop a GPU-based solution that (i) considerably differs from existing CPU-based methods, and (ii) exploits the strengths of GPUs in terms of parallelism while mitigating their limitations in terms of the computation power per GPU core. With extensive experiments on a variety of biological networks, we show that our solution is up to two orders of magnitude faster than the best CPU-based approach, and is around 20 times more cost-effective than the latter, when taking into account the monetary costs of the CPU and GPUs used.

Driven by the advances in location positioning techniques and the popularity of location sharing services, semantic enriched trajectory data have become unprecedentedly available. While finding relevant Point-of-Interest (POIs) based on users' locations and query keywords has been extensively studied in the past years, it is largely untouched to explore the keyword queries in the context of semantic trajectory database. In this paper, we study the problem of approximate keyword search in massive semantic trajectories. Given a set of query keywords, an approximate keyword query of semantic trajectory (AKQST) returns k trajectories that contain the most relevant keywords to the query and yield the least travel effort in the meantime. The main difference between AKQST and conventional spatial keyword queries is that there is no query location in AKQST, which means the search area cannot be localized. To capture the travel effort in the context of query keywords, a novel utility function, called spatio-textual utility function, is first defined. Then we develop a hybrid index structure called GiKi to organize the trajectories hierarchically, which enables pruning the search space by spatial and textual similarity simultaneously. Finally an efficient search algorithm and fast evaluation of the minimum value of spatio-textual utility function are proposed. The results of our empirical studies based on real check-in datasets demonstrate that our proposed index and algorithms can achieve good scalability.

Location recommendation plays an essential role in helping people find places they are likely to enjoy. Though some recent research has studied how to recommend locations with the presence of social network and geographical information, few of them addressed the cold-start problem, specifically, recommending locations for new users. Because the visits to locations are often shared on social networks, rich semantics (e.g., tweets) that reveal a person's interests can be leveraged to tackle this challenge. A typical way is to feed them into traditional explicit-feedback content-aware recommendation methods (e.g., LibFM). As a user's negative preferences are not explicitly observable in most human mobility data, these methods need draw negative samples for better learning performance. However, prior studies have empirically shown that sampling-based methods don't perform as well as a method that considers all unvisited locations as negative but assigns them a lower confidence. To this end, we propose an Implicit-feedback based Content-aware Collaborative Filtering (ICCF) framework to incorporate semantic content and steer clear of negative sampling. For efficient parameter learning, we develop a scalable optimization algorithm, scaling linearly with the data size and the feature size. Furthermore, we offer a good explanation to ICCF, such that the semantic content is actually used to refine user similarity based on mobility. Finally, we evaluate ICCF with a large-scale LBSN dataset where users have profiles and text content. The results show that ICCF outperforms LibFM of the best configuration, and that user profiles and text content are not only effective at improving recommendation but also helpful for coping with the cold-start problem.

In the retail market, the consumers' indecisiveness refers to the inability to make quick and assertive decisions when they choose among competing product options. Indeed, indecisiveness has been investigated in a number of fields, such as economics and psychology. However, these studies are usually based on the subjective customer survey data with some manually defined questions. Instead, in this paper, we provide a focused study on automatically mining indecisiveness in massive customer behaviors in online stores. Specifically, we first give a general definition to measure the observed indecisiveness in each behavior session. From these observed indecisiveness, we can learn the latent factors/reasons by a probabilistic factor-based model. These two factors are the indecisive indexes of the customers and the product bundles, respectively. Next, we demonstrate that this indecisiveness mining process could be useful in several potential applications, such as the competitive product detection and personalized product bundles recommendation. Finally, we perform extensive experiments on a large-scale behavioral logs of online customers in a distributed environment. The results reveal that our measurement of indecisiveness agrees with the common sense assessment, and the discoveries are useful in predicting customer behaviors and providing better recommendation services for both customers and online retailers.

Mobility prediction enables appealing proactive experiences for location-aware services and offers essential intelligence to business and governments. Recent studies suggest that human mobility is highly regular and predictable. Additionally, social conformity theory indicates that people's movements are influenced by others. However, existing approaches for location prediction fail to organically combine both the regularity and conformity of human mobility in a unified model, and lack the capacity to incorporate heterogeneous mobility datasets to boost prediction performance. To address these challenges, in this paper we propose a hybrid predictive model integrating both the regularity and conformity of human mobility as well as their mutual reinforcement. In addition, we further elevate the predictive power of our model by learning location profiles from heterogeneous mobility datasets based on a gravity model. We evaluate the proposed model using several city-scale mobility datasets including location check-ins, GPS trajectories of taxis, and public transit data. The experimental results validate that our model significantly outperforms state-of-the-art approaches for mobility prediction in terms of multiple metrics such as accuracy and percentile rank. The results also suggest that the predictability of human mobility is time-varying, e.g., the overall predictability is higher on workdays than holidays while predicting users' unvisited locations is more challenging for workdays than holidays.

As the worlds of commerce and Internet technology become more inextricably linked, a large number of user consumption series become available for creative use. A critical demand along this line is to predict the future product adoption for the merchants, which enables a wide range of applications such as targeted marketing. However, previous works only aimed at predicting if one user will adopt this product or not; the problem of adoption rate (or percentage of use) prediction for each user is still underexplored due to the complexity of user decision-making process. To that end, in this paper we present a comprehensive study for this product adoption rate prediction problem. Specifically, we first introduce a decision function to capture the change of users' product adoption rate, where various factors that may influence the decision can be generally leveraged. Then, we propose two models to solve this function, the Generalized Adoption Model (GAM) that assumes all users are influenced equally by these factors and the Personalized Adoption Model (PAM) that argues each factor contributes differently among people. Furthermore, we extend the PAM to a totally Bayesian model (BPAM) that can automatically learn all parameters. Finally, extensive experiments on two real-world datasets not only show the improvement of our proposed three models, but also give insights to track the effects of the various factors for product adoption decisions.

User profiling is crucial to many online services. Several recent studies suggest that demographic attributes are predictable from different online behavioral data, such as users' "Likes" on Facebook, friendship relations, and the linguistic characteristics of tweets. But location check-ins, as a bridge of users' offline and online lives, have by and large been overlooked in inferring user profiles. In this paper, we investigate the predictive power of location check-ins for inferring users' demographics and propose a simple yet general location to profile (L2P) framework. More specifically, we extract rich semantics of users' check-ins in terms of spatiality, temporality, and location knowledge, where the location knowledge is enriched with semantics mined from heterogeneous domains including both online customer review sites and social networks. Additionally, tensor factorization is employed to draw out low dimensional representations of users' intrinsic check-in preferences considering the above factors. Meanwhile, the extracted features are used to train predictive models for inferring various demographic attributes.

The rapid growth of location-based services provide the potential to understand people's mobility pattern at an unprecedented level, which can also enable food-service industry to accurately predict consumer's dining behavior. In this paper, by leveraging users' historical dining pattern, socio-demographic characteristics and restaurants' attributes, we aim at generating the top-K restaurants for a user's next dining. Compared to previous studies in location prediction which mainly focus on regular mobility patterns, we present a novelty-seeking based dining recommender system, termed NDRS, in consideration of both exploration and exploitation. First, we apply a Conditional Random Field (CRF) with additional constraints to infer users' novelty-seeking statuses by considering both spatial-temporal-historical features and users' socio-demographic characteristics. On the one hand, when a user is predicted to be novelty-seeking, by incorporating the influence of restaurants' contextual factors such as price and service quality, we propose a context-aware collaborative filtering method to recommend restaurants she has never visited before. On the other hand, when a user is predicted to be not novelty-seeking, we then present a Hidden Markov Model (HMM) considering the temporal regularity to recommend the previously visited restaurants. To evaluate the performance of each component as well as the whole system, we conduct extensive experiments, with a large dataset we have collected covering the concerned dining related check-ins, users' demographics, and restaurants' attributes. The results reveal that our system is effective for dining recommendation.

The papers in this special section provide the opportunity for researchers, practitioners, and application developers to review and discuss the state-of-the-art and trends of MCS (mobile crowd sensing)techniques and applications or propose new solutions. The ubiquitous sensor-rich mobile devices (e.g., smartphones, wearable devices, and smart vehicles) have been playing an increasing important role in the evolution of the Internet of Things (IoTs), which bridges the digital space and physical space at a societal scale. Their powerful computing/communication capacities, huge population, and inherent mobility make mobile-device networks a much more flexible and cost-effective IoT solution than static sensor networks.

SimRank is an important measure of vertex-pair similarity according to the structure of graphs. The similarity search based on SimRank is an important operation for identifying similar vertices in a graph and has been employed in many data analysis applications. Nowadays, graphs in the real world become much larger and more dynamic. The existing solutions for similarity search are expensive in terms of time and space cost. None of them can efficiently support similarity search over large dynamic graphs. In this paper, we propose a novel two-stage random-walk sampling framework (TSF) for SimRank-based similarity search (e.g., top- k search). In the preprocessing stage, TSF samples a set of one-way graphs to index raw random walks in a novel manner within O ( NR g ) time and space, where N is the number of vertices and R g is the number of one-way graphs. The one-way graph can be efficiently updated in accordance with the graph modification, thus TSF is well suited to dynamic graphs. During the query stage, TSF can search similar vertices fast by naturally pruning unqualified vertices based on the connectivity of one-way graphs. Furthermore, with additional R q samples, TSF can estimate the SimRank score with probability [EQUATION] if the error of approximation is bounded by 1 -- ε. Finally, to guarantee the scalability of TSF, the one-way graphs can also be compactly stored on the disk when the memory is limited. Extensive experiments have demonstrated that TSF can handle dynamic billion-edge graphs with high performance.

The study of online social networks has attracted increasing interest. However, concerns are raised for the privacy risks of user data since they have been frequently shared among researchers, advertisers, and application developers. To solve this problem, a number of anonymization algorithms have been recently developed for protecting the privacy of social graphs. In this article, we proposed a graph node similarity measurement in consideration with both graph structure and descriptive information, and a deanonymization algorithm based on the measurement. Using the proposed algorithm, we evaluated the privacy risks of several typical anonymization algorithms on social graphs with thousands of nodes from Microsoft Academic Search, LiveJournal, and the Enron email dataset, and a social graph with millions of nodes from Tencent Weibo. Our results showed that the proposed algorithm was efficient and effective to deanonymize social graphs without any initial seed mappings. Based on the experiments, we also pointed out suggestions on how to better maintain the data utility while preserving privacy.

Coming with the popularity of multimedia sharing platforms such as Facebook and Flickr, recent years have witnessed an explosive growth of geographical tags on social multimedia content. This trend enables a wide variety of emerging applications, for example, mobile location search, landmark recognition, scene reconstruction, and touristic recommendation, which range from purely research prototype to commercial systems. In this article, we give a comprehensive survey on these applications, covering recent advances in recognition and mining of geographical-aware social multimedia. We review related work in the past decade regarding to location recognition, scene summarization, tourism suggestion, 3D building modeling, mobile visual search and city navigation. At the end, we further discuss potential challenges, future topics, as well as open issues related to geo-social multimedia computing, recognition, mining, and analytics.

Urban transportation is an important factor in energy consumption and pollution, and is of increasing concern due to its complexity and economic significance. Its importance will only increase as urbanization continues around the world. In this article, we explore drivers’ refueling behavior in urban areas. Compared to questionnaire-based methods of the past, we propose a complete data-driven system that pushes towards real-time sensing of individual refueling behavior and citywide petrol consumption. Our system provides the following: detection of individual refueling events (REs) from which refueling preference can be analyzed; estimates of gas station wait times from which recommendations can be made; an indication of overall fuel demand from which macroscale economic decisions can be made, and a spatial, temporal, and economic view of urban refueling characteristics. For individual behavior, we use reported trajectories from a fleet of GPS-equipped taxicabs to detect gas station visits. For time spent estimates, to solve the sparsity issue along time and stations, we propose context-aware tensor factorization (CATF), a factorization model that considers a variety of contextual factors (e.g., price, brand, and weather condition) that affect consumers’ refueling decision. For fuel demand estimates, we apply a queue model to calculate the overall visits based on the time spent inside the station. We evaluated our system on large-scale and real-world datasets, which contain 4-month trajectories of 32,476 taxicabs, 689 gas stations, and the self-reported refueling details of 8,326 online users. The results show that our system can determine REs with an accuracy of more than 90%, estimate time spent with less than 2 minutes of error, and measure overall visits in the same order of magnitude with the records in the field study.

The step of urbanization and modern civilization fosters different functional zones in a city, such as residential areas, business districts, and educational areas. In a metropolis, people commute between these functional zones every day to engage in different socioeconomic activities, e.g., working, shopping, and entertaining. In this paper, we propose a data-driven framework to discover functional zones in a city. Specifically, we introduce the concept of latent activity trajectory (LAT), which captures socioeconomic activities conducted by citizens at different locations in a chronological order. Later, we segment an urban area into disjointed regions according to major roads, such as highways and urban expressways. We have developed a topic-modeling-based approach to cluster the segmented regions into functional zones leveraging mobility and location semantics mined from LAT. Furthermore, we identify the intensity of each functional zone using Kernel Density Estimation. Extensive experiments are conducted with several urban scale datasets to show that the proposed framework offers a powerful ability to capture city dynamics and provides valuable calibrations to urban planners in terms of functional zones.

Community bulletin boards serve an important function for public information sharing in modern society. Posted fliers advertise services, events, and other announcements. However, fliers posted offline suffer from problems such as limited spatial-temporal coverage and inefficient search support. In recent years, with the development of sensor-enhanced mobile devices, mobile crowd sensing (MCS) has been used in a variety of application areas. This paper presents FlierMeet, a crowd-powered sensing system for cross-space public information reposting, tagging, and sharing. The tags learned are useful for flier sharing and preferred information retrieval and suggestion. Specifically, we utilize various contexts (e.g., spatio-temporal info, flier publishing/reposting behaviors, etc.) and textual features to group similar reposts and classify them into categories. We further identify a novel set of crowd-object interaction hints to predict the semantic tags of reposts. To evaluate our system, 38 participants were recruited and 2,035 reposts were captured during an eight-week period. Experiments on this dataset showed that our approach to flier grouping is effective and the proposed features are useful for flier category/semantic tagging.

Anonymized user datasets are often released for research or indus-try applications. As an example, t.qq.com released its anonymized users ’ profile, social interaction, and recommendation log data in KDD Cup 2012 to call for recommendation algorithms. Since the entities (users and so on) and edges (links among entities) are of multiple types, the released social network is a heterogeneous in-formation network. Prior work has shown how privacy can be com-promised in homogeneous information networks by the use of spe-cific types of graph patterns. We show how the extra information derived from heterogeneity can be used to relax these assumptions. To characterize and demonstrate this added threat, we formally de-fine privacy risk in an anonymized heterogeneous information net-work to identify the vulnerability in the possible way such data are released, and further present a new de-anonymization attack that exploits the vulnerability. Our attack successfully de-anonymized most individuals involved in the data—for an anonymized 1,000-user t.qq.com network of density 0.01, the attack precision is over 90 % with a 2.3-million-user auxiliary network.

The emergence of social networks has provided opportunities for both targeted marketing and viral marketing. By concentrating the efforts on a few key customers, targeted marketing could make the promotion of the items (products) much easier and more cost-effective. On the other hand, viral marketing aims at finding a set of individuals (seeds) to maximize the word-of-mouth propagation of an item. However, these two marketing strategies can only exploit some specific characteristics of the social networks, and the problem of how to combine them together to build a better, stronger business is still open. To that end, in this paper, we propose a general approach for integrated marketing. Specifically, to market a given item, we first generate the item-specific candidate users by a recommendation algorithm, and then select the typical users who have the best balanced utility scores and consumption/social entropy. Next, treating typical users as targeted customers, we study the problem of maximizing information awareness in viral marketing with these constrained targets. Along this line, we define it as a constrained coverage maximization problem, and propose three solutions: GMIC, LMIC and QMIC. Finally, extensive experimental results on real-world datasets demonstrate that our integrated marketing approach could outperform the methods that consider only targeted marketing or viral marketing.

Point-of-Interest (POI) recommendation has become an important means to help people discover attractive locations. However, extreme sparsity of user-POI matrices creates a severe challenge. To cope with this challenge, viewing mobility records on location-based social networks (LBSNs) as implicit feedback for POI recommendation, we first propose to exploit weighted matrix factorization for this task since it usually serves collaborative filtering with implicit feedback better. Besides, researchers have recently discovered a spatial clustering phenomenon in human mobility behavior on the LBSNs, i.e., individual visiting locations tend to cluster together, and also demonstrated its effectiveness in POI recommendation, thus we incorporate it into the factorization model. Particularly, we augment users' and POIs' latent factors in the factorization model with activity area vectors of users and influence area vectors of POIs, respectively. Based on such an augmented model, we not only capture the spatial clustering phenomenon in terms of two-dimensional kernel density estimation, but we also explain why the introduction of such a phenomenon into matrix factorization helps to deal with the challenge from matrix sparsity. We then evaluate the proposed algorithm on a large-scale LBSN dataset. The results indicate that weighted matrix factorization is superior to other forms of factorization models and that incorporating the spatial clustering phenomenon into matrix factorization improves recommendation performance.

An incisive understanding of personal psychological traits is not only essential to many scientific disciplines, but also has a profound business impact on online recommendation. Recent studies in psychology suggest that novelty-seeking trait is highly related to consumer behavior. In this paper, we focus on understanding individual novelty-seeking trait embodied at different levels and across heterogeneous domains. Unlike the questionnaire-based methods widely adopted in the past, we first present a computational framework, Novel Seeking Model (NSM), for exploring the novelty-seeking trait implied by observable activities. Then, we explore the novelty-seeking trait in two heterogeneous domains: check-in behavior in location based social networks, which reflects mobility patterns in the physical world, and online shopping behavior on e-commerce sites, which reflects consumption concepts in economic activities. To demonstrate the effectiveness of NSM, we conducted extensive experiments, with a large dataset covering the two-domain activities for hundreds of thousands of individuals. Our results suggest that NSM offers a powerful paradigm for 1) presenting an effective measurement of a personality trait that can explicitly explain the deviation of individuals from the habits of individuals and crowds; 2) uncovering the correlation of novelty-seeking trait at different levels and across heterogeneous domains. The proposed method provides emerging implications for personalized cross-domain recommendation and targeted advertising.

Retrieving timely and relevant information on-site is an important task for mobile users. A context-aware system can understand a user’s information needs and thus select contents according to relevance. We propose a context-dependent search engine that represents user context in a knowledge-based context model, implemented in a hierarchical structure with granularity information. Search results are ordered based on semantic relevance computed as similarity between the current context and tags of search results. Compared against baseline algorithms, the proposed approach enhances precision by 22% and pooled recall by 17%. The use of size-based granularity to compute similarity makes the approach more robust against changes in the context model in comparison to graph-based methods, facilitating import of existing knowledge repositories and end-user defined vocabularies (folksonomies). The reasoning engine being light-weight, privacy protection is ensured, as all user information is processed locally on the user’s phone without requiring communication with an external server.

Many innovative location-based services have been established to offer users greater convenience in their everyday lives. These services usually cannot map user's physical locations into semantic names automatically. The semantic names of locations provide important context for mobile recommendations and advertisements. In this article, we proposed a novel location naming approach which can automatically provide semantic names for users given their locations and time. In particular, when a user opens a GPS device and submits a query with her physical location and time, she will be returned the most appropriate semantic name. In our approach, we drew an analogy between location naming and local search, and designed a local search framework to propose a spatiotemporal and user preference (STUP) model for location naming. STUP combined three components, user preference (UP), spatial preference (SP), and temporal preference (TP), by leveraging learning-to-rank techniques. We evaluated STUP on 466,190 check-ins of 5,805 users from Shanghai and 135,052 check-ins of 1,361 users from Beijing. The results showed that SP was most effective among three components and that UP can provide personalized semantic names, and thus it was a necessity for location naming. Although TP was not as discriminative as the others, it can still be beneficial when integrated with SP and UP. Finally, according to the experimental results, STUP outperformed the proposed baselines and returned accurate semantic names for 23.6% and 26.6% of the testing queries from Beijing and Shanghai, respectively.

Urban transportation is increasingly studied due to its complexity and economic importance. It is also a major component of urban energy use and pollution. The importance of this topic will only increase as urbanization continues around the world. A less researched aspect of transportation is the refueling behavior of drivers. In this paper, we propose a step toward real-time sensing of refueling behavior and citywide petrol consumption. We use reported trajectories from a fleet of GPS-equipped taxicabs to detect gas station visits, measure the time spent, and estimate overall demand. For times and stations with sparse data, we use collaborative filtering to estimate conditions. Our system provides real-time estimates of gas stations' waiting times, from which recommendations could be made, an indicator of overall gas usage, from which macro-scale economic decisions could be made, and a geographic view of the efficiency of gas station placement.

Destination prediction is an essential task for many emerging location based applications such as recommending sightseeing places and targeted advertising based on destination. A common approach to destination prediction is to derive the probability of a location being the destination based on historical trajectories. However, existing techniques using this approach suffer from the “data sparsity problem”, i.e., the available historical trajectories is far from being able to cover all possible trajectories. This problem considerably limits the number of query trajectories that can obtain predicted destinations. We propose a novel method named Sub-Trajectory Synthesis (SubSyn) algorithm to address the data sparsity problem. SubSyn algorithm first decomposes historical trajectories into sub-trajectories comprising two neighbouring locations, and then connects the sub-trajectories into “synthesised” trajectories. The number of query trajectories that can have predicted destinations is exponentially increased by this means. Experiments based on real datasets show that SubSyn algorithm can predict destinations for up to ten times more query trajectories than a baseline algorithm while the SubSyn prediction algorithm runs over two orders of magnitude faster than the baseline algorithm. In this paper, we also consider the privacy protection issue in case an adversary uses SubSyn algorithm to derive sensitive location information of users. We propose an efficient algorithm to select a minimum number of locations a user has to hide on her trajectory in order to avoid privacy leak. Experiments also validate the high efficiency of the privacy protection algorithm.

Smart card transactions capture rich information of human mobility and urban dynamics, therefore are of particular interest to urban planners and location-based service providers. However, since most transaction systems are only designated for billing purpose, typically, fine-grained location information, such as the exact boarding and alighting stops of a bus trip, is only partially or not available at all, which blocks deep exploitation of this rich and valuable data at individual level. This paper presents a "space alignment" framework to reconstruct individual mobility history from a large-scale smart card transaction dataset pertaining to a metropolitan city. Specifically, we show that by delicately aligning the monetary space and geospatial space with the temporal space, we are able to extrapolate a series of critical domain specific constraints. Later, these constraints are naturally incorporated into a semi-supervised conditional random field to infer the exact boarding and alighting stops of all transit routes with a surprisingly high accuracy, e.g., given only 10% trips with known alighting/boarding stops, we successfully inferred more than 78% alighting and boarding stops from all unlabeled trips. In addition, we demonstrated that the smart card data enriched by the proposed approach dramatically improved the performance of a conventional method for identifying users' home and work places (with 88% improvement on home detection and 35% improvement on work place detection). The proposed method offers the possibility to mine individual mobility from common public transit transactions, and showcases how uncertain data can be leveraged with domain knowledge and constraints, to support cross-application data mining tasks.

Destination prediction is an essential task for a number of emerging location based applications such as recommending sightseeing places and sending targeted advertisements. A common approach to destination prediction is to derive the probability of a location being the destination based on historical trajectories. However, existing techniques suffer from the "data sparsity problem", i.e., the number of available historical trajectories is far from sufficient to cover all possible trajectories. This problem considerably limits the amount of query trajectories whose predicted destinations can be inferred. In this demonstration, we showcase a system named "DesTeller" that is interactive, user-friendly, publicly accessible, and capable of answering real-time queries. The underlying algorithm Sub-Trajectory Synthesis (SubSyn) successfully addressed the data sparsity problem and is able to predict destinations for almost every query submitted by travellers. We also consider the privacy protection issue in case an adversary uses SubSyn algorithm to derive sensitive location information of users.

This paper presents a smart driving direction system leveraging the intelligence of experienced drivers. In this system, GPS-equipped taxis are employed as mobile sensors probing the traffic rhythm of a city and taxi drivers' intelligence in choosing driving directions in the physical world. We propose a time-dependent landmark graph to model the dynamic traffic pattern as well as the intelligence of experienced drivers so as to provide a user with the practically fastest route to a given destination at a given departure time. Then, a Variance-Entropy-Based Clustering approach is devised to estimate the distribution of travel time between two landmarks in different time slots. Based on this graph, we design a two-stage routing algorithm to compute the practically fastest and customized route for end users. We build our system based on a real-world trajectory data set generated by over 33,000 taxis in a period of three months, and evaluate the system by conducting both synthetic experiments and in-the-field evaluations. As a result, 60-70 percent of the routes suggested by our method are faster than the competing methods, and 20 percent of the routes share the same results. On average, 50 percent of our routes are at least 20 percent faster than the competing approaches.

This paper presents a recommender system for both taxi drivers and people expecting to take a taxi, using the knowledge of 1) passengers' mobility patterns and 2) taxi drivers' picking-up/dropping-off behaviors learned from the GPS trajectories of taxicabs. First, this recommender system provides taxi drivers with some locations and the routes to these locations, toward which they are more likely to pick up passengers quickly (during the routes or in these locations) and maximize the profit of the next trip. Second, it recommends people with some locations (within a walking distance) where they can easily find vacant taxis. In our method, we learn the above-mentioned knowledge (represented by probabilities) from GPS trajectories of taxis. Then, we feed the knowledge into a probabilistic model that estimates the profit of the candidate locations for a particular driver based on where and when the driver requests the recommendation. We build our system using historical trajectories generated by over 12,000 taxis during 110 days and validate the system with extensive evaluations including in-the-field user studies.

A variety of services utilizing users' positions have become available because of rapid advances in Global Positioning System (GPS) technologies. Since location information may reveal private information, preserving location privacy has become a significant issue. We proposed a dummy-based method of anonymizing location to protect this privacy in our previous work that generated dummies based on various restrictions in a real environment. However, the previous work assumed a simplified mobility model in which users kept moving and did not stop. If we assume a more realistic mobility model in which users often pause to visit various attractions, it becomes increasingly more difficult to generate dummies that will move naturally. In this paper, we assumed that the users' movements are known in advance and propose a dummy-based anonymization method based on user movements, where dummies move naturally while stopping at several locations. We simulated user movements on real map information and verified the method we propose was more effective than the previous one.

The increasing availability of GPS-embedded mobile devices has given rise to a new spectrum of location-based services, which have accumulated a huge collection of location trajectories. In practice, a large portion of these trajectories are of low-sampling-rate. For instance, the time interval between consecutive GPS points of some trajectories can be several minutes or even hours. With such a low sampling rate, most details of their movement are lost, which makes them difficult to process effectively. In this work, we investigate how to reduce the uncertainty in such kind of trajectories. Specifically, given a low-sampling-rate trajectory, we aim to infer its possible routes. The methodology adopted in our work is to take full advantage of the rich information extracted from the historical trajectories. We propose a systematic solution, History based Route Inference System (HRIS), which covers a series of novel algorithms that can derive the travel pattern from historical data and incorporate it into the route inference process. To validate the effectiveness of the system, we apply our solution to the map-matching problem which is an important application scenario of this work, and conduct extensive experiments on a real taxi trajectory dataset. The experiment results demonstrate that HRIS can achieve higher accuracy than the existing map-matching algorithms for low-sampling-rate trajectories.

The development of a city gradually fosters different functional regions, such as educational areas and business districts. In this paper, we propose a framework (titled DRoF) that Discovers Regions of different Functions in a city using both human mobility among regions and points of interests (POIs) located in a region. Specifically, we segment a city into disjointed regions according to major roads, such as highways and urban express ways. We infer the functions of each region using a topic-based inference model, which regards a region as a document, a function as a topic, categories of POIs (e.g., restaurants and shopping malls) as metadata (like authors, affiliations, and key words), and human mobility patterns (when people reach/leave a region and where people come from and leave for) as words. As a result, a region is represented by a distribution of functions, and a function is featured by a distribution of mobility patterns. We further identify the intensity of each function in different locations. The results generated by our framework can benefit a variety of applications, including urban planning, location choosing for a business, and social recommendations. We evaluated our method using large-scale and real-world datasets, consisting of two POI datasets of Beijing (in 2010 and 2011) and two 3-month GPS trajectory datasets (representing human mobility) generated by over 12,000 taxicabs in Beijing in 2010 and 2011 respectively. The results justify the advantages of our approach over baseline methods solely using POIs or human mobility.

Yellow pages search is a popular service that provides a means for finding businesses close to particular locations. The efficient search of yellow pages is becoming a rapidly evolving research area. The underlying data maintained in yellow pages search engines are typically labeled according to Standard Industry Classification (SIC) categories, and users can search yellow pages with categories according to their interests. Categorizing yellow pages queries into a subset of topical categories can help to improve search experience and quality. However, yellow pages queries are usually short and ambiguous. In addition, a yellow pages query taxonomy is typically organized by a hierarchy of a fairly large number of categories. These characteristics make automatic yellow pages query categorization difficult and challenging. In this article, we propose a flexible yellow pages query categorization approach. The proposed technique is built based on a TF-IDF similarity taxonomy matching scheme that is able to provide more accurate query categorization than previous keyword-based matching schemes. To further improve the categorization performance, we design several filtering schemes. Through extensive experimentation, we demonstrate encouraging results. We obtain F1 measures of about 0.5 and 0.3 for categorizing yellow pages queries into 19 coarse categories and 244 finer categories, respectively. We investigate different components in the proposed approach and also demonstrate the superiority of our approach over a hierarchical support vector machine classifier.

We propose a flexible mini-shuttle like transportation system called flexi, with routes formed by analyzing passenger trip data from a large set of taxi trajectories. The usage of public transportation is declining as often it no longer matches with individual needs. Thus, the flexi system provides a transportation mode in between buses and taxis so that inconvenience in switching to the system can be minimized overall. To generate flexi routes, we propose a two-phase approach. In the first phase, a fast diameter-constrained agglomerative clustering algorithm is developed and applied to the set of trips derived from the GPS data. This phase identifies a set of heavily traveled spatio-temporal trip clusters called hot lines. In the second phase, a directed acyclic graph is constructed from the hot lines. Then, an optimal single flexi route discovery algorithm on graph searching is proposed. Multiple routes are discovered by iteratively applying the single routing algorithm. Extensive experiments using a large set of real taxi trajectory data show that the flexi system can save a large percentage of trip mileage.

Many innovative location-based services have been established in order to facilitate users' everyday lives. Usually, these services cannot obtain location names automatically from users' GPS coordinates to claim their current locations. In this paper, we propose a novel location naming approach, which can provide concrete and meaningful location names to users based on their current location, time and check-in histories. In particular, when users input a GPS point, they will receive a ranked list of Points of Interest which shows the most possible semantic names for that location. In our approach, we draw an analogy between the location naming problem and the location-based search problem. We proposed a local search framework to integrate different kinds of popularity factors and personal preferences. After identifying important features by feature selection, we apply learning-to-rank technique to weight them and build our system based on 31811 check-in records from 545 users. By evaluating on this dataset, our approach is shown to be effective in automatically naming users' locations. 64.5% of test queries can return the intended location names within the top 5 results.

No abstract available.

We present a recommender for taxi drivers and people expecting to take a taxi, using the knowledge of 1) passengers' mobility patterns and 2) taxi drivers' pick-up behaviors learned from the GPS trajectories of taxicabs. First, this recommender provides taxi drivers with some locations and the routes to these locations, towards which they are more likely to pick up passengers quickly (during the routes or at these locations) and maximize the profit. Second, it recommends people with some locations (within a walking distance) where they can easily find vacant taxis. In our method, we learn the above knowledge (represented by probabilities) from GPS trajectories of taxis. Then, we feed the knowledge into a probabilistic model which estimates the profit of the candidate locations for a particular driver based on where and when the driver requests for the recommendation. We validate our recommender using historical trajectories generated by over 12,000 taxis during 110 days.

Urban computing for city planning is one of the most significant applications in Ubiquitous computing. In this paper we detect flawed urban planning using the GPS trajectories of taxicabs traveling in urban areas. The detected results consist of 1) pairs of regions with salient traffic problems and 2) the linking structure as well as correlation among them. These results can evaluate the effectiveness of the carried out planning, such as a newly built road and subway lines in a city, and remind city planners of a problem that has not been recognized when they conceive future plans. We conduct our method using the trajectories generated by 30,000 taxis from March to May in 2009 and 2010 in Beijing, and evaluate our results with the real urban planning of Beijing.

The detection of outliers in spatio-temporal traffic data is an important research problem in the data mining and knowledge discovery community. However to the best of our knowledge, the discovery of relationships, especially causal interactions, among detected traffic outliers has not been investigated before. In this paper we propose algorithms which construct outlier causality trees based on temporal and spatial properties of detected outliers. Frequent substructures of these causality trees reveal not only recurring interactions among spatio-temporal outliers, but potential flaws in the design of existing traffic networks. The effectiveness and strength of our algorithms are validated by experiments on a very large volume of real taxi trajectories in an urban road network.

This paper presents a Cloud-based system computing customized and practically fast driving routes for an end user using (historical and real-time) traffic conditions and driver behavior. In this system, GPS-equipped taxicabs are employed as mobile sensors constantly probing the traffic rhythm of a city and taxi drivers' intelligence in choosing driving directions in the physical world. Meanwhile, a Cloud aggregates and mines the information from these taxis and other sources from the Internet, like Web maps and weather forecast. The Cloud builds a model incorporating day of the week, time of day, weather conditions, and individual driving strategies (both of the taxi drivers and of the end user for whom the route is being computed). Using this model, our system predicts the traffic conditions of a future time (when the computed route is actually driven) and performs a self-adaptive driving direction service for a particular user. This service gradually learns a user's driving behavior from the user's GPS logs and customizes the fastest route for the user with the help of the Cloud. We evaluate our service using a real-world dataset generated by over 33,000 taxis over a period of 3 months in Beijing. As a result, our service accurately estimates the travel time of a route for a user; hence finding the fastest route customized for the user.

In this paper, we aim to develop a travelogue service that discovers and conveys various travelogue digests, in form of theme locations, geographical scope, traveling trajectory and location snippet, to users. In this service, theme locations in a travelogue are the core information to discover. Thus we aim to address the problem of theme location discovery to enable the above travelogue services. Due to the inherent ambiguity of location relevance, we perform location relevance mining (LRM) in two complementary angles, relevance classification and relevance ranking, to provide comprehensive understanding of locations. Furthermore, we explore the textual (e.g., surrounding words) and geographical (e.g., geographical relationship among locations) features of locations to develop a co-training model for enhancement of classification performance. Built upon the mining result of LRM, we develop a series of techniques for provisioning of the aforementioned travelogue digests in our travelogue system. Finally, we conduct comprehensive experiments on collected travelogues to evaluate the performance of our location relevance mining techniques and demonstrate the effectiveness of the travelogue service.

The advance of GPS-enabled devices allows people to record their location histories with GPS traces, which imply human behaviors and preferences related to travel. In this article, we perform two types of travel recommendations by mining multiple users' GPS traces. The first is a generic one that recommends a user with top interesting locations and travel sequences in a given geospatial region. The second is a personalized recommendation that provides an individual with locations matching her travel preferences. To achieve the first recommendation, we model multiple users' location histories with a tree-based hierarchical graph ( TBHG ). Based on the TBHG , we propose a HITS (Hypertext Induced Topic Search)-based model to infer the interest level of a location and a user's travel experience (knowledge). In the personalized recommendation, we first understand the correlation between locations, and then incorporate this correlation into a collaborative filtering (CF)-based model, which predicts a user's interests in an unvisited location based on her locations histories and that of others. We evaluated our system based on a real-world GPS trace dataset collected by 107 users over a period of one year. As a result, our HITS-based inference model outperformed baseline approaches like rank-by-count and rank-by-frequency . Meanwhile, we achieved a better performance in recommending travel sequences beyond baselines like rank-by-count . Regarding the personalized recommendation, our approach is more effective than the weighted Slope One algorithm with a slightly additional computation, and is more efficient than the Pearson correlation-based CF model with the similar effectiveness.

The increasing availability of location-acquisition technologies (GPS, GSM networks, etc.) enables people to log the location histories with spatio-temporal data. Such real-world location histories imply, to some extent, users' interests in places, and bring us opportunities to understand the correlation between users and locations. In this article, we move towards this direction and report on a personalized friend and location recommender for the geographical information systems (GIS) on the Web. First, in this recommender system, a particular individual's visits to a geospatial region in the real world are used as their implicit ratings on that region. Second, we measure the similarity between users in terms of their location histories and recommend to each user a group of potential friends in a GIS community. Third, we estimate an individual's interests in a set of unvisited regions by involving his/her location history and those of other users. Some unvisited locations that might match their tastes can be recommended to the individual. A framework, referred to as a hierarchical-graph-based similarity measurement (HGSM), is proposed to uniformly model each individual's location history, and effectively measure the similarity among users. In this framework, we take into account three factors: 1) the sequence property of people's outdoor movements, 2) the visited popularity of a geospatial region, and 3) the hierarchical property of geographic spaces. Further, we incorporated a content-based method into a user-based collaborative filtering algorithm, which uses HGSM as the user similarity measure, to estimate the rating of a user on an item. We evaluated this recommender system based on the GPS data collected by 75 subjects over a period of 1 year in the real world. As a result, HGSM outperforms related similarity measures, namely similarity-by-count, cosine similarity, and Pearson similarity measures. Moreover, beyond the item-based CF method and random recommendations, our system provides users with more attractive locations and better user experiences of recommendation.

With the increasing popularity of location tracking services such as GPS, more and more mobile data are being accumulated. Based on such data, a potentially useful service is to make timely and targeted recommendations for users on places where they might be interested to go and activities that they are likely to conduct. For example, a user arriving in Beijing might wonder where to visit and what she can do around the Forbidden City. A key challenge for such recommendation problems is that the data we have on each individual user might be very limited, while to make useful and accurate recommendations, we need extensive annotated location and activity information from user trace data. In this paper, we present a new approach, known as user-centered collaborative location and activity filtering (UCLAF), to pull many users’ data together and apply collaborative filtering to find like-minded users and like-patterned activities at different locations. We model the userlocation- activity relations with a tensor representation, and propose a regularized tensor and matrix decomposition solution which can better address the sparse data problem in mobile information retrieval. We empirically evaluate UCLAF using a real-world GPS dataset collected from 164 users over 2.5 years, and showed that our system can outperform several state-of-the-art solutions to the problem.

A typical location extraction approach consists of two steps, location name detection and location entity disambiguation. Promising results have been obtained in the last decade based on natural language processing technologies. However, there are still two challenges which requires further investigation: 1)How to leverage the prior and contextual evidence to improve the location extraction performance, and 2) How to utilize the interdependence information between the named entity recognition step and disambiguation step. In this paper, we propose an iterative detection-ranking framework to address these problems as well as a set of novel features to mine contextual information from web resources. Experimental results show that our solution outperforms the state-of-the-art approaches, including Metacarta GeoTagger and Yahoo Placemaker.

Recent mobile devices are mostly equipped with a GPS receiver. This trend has arisen a variety of location based services (LBSs) which enable users to search local information of their current locations. To use LBSs, a user needs to send his/her location information to a service provider. Users' location information is inherently private, since it can reveal the critical information. In this paper, we propose a method to protect the user's location privacy by sending the user's location with dummy locations, which are determined based on the user's current location. We conduct an experiment to evaluate the effectiveness of our proposed method using real users' trajectories on a real map.

In this paper, we aim to estimate the similarity between users according to their GPS trajectories. Our approach first models a user's GPS trajectories with a semantic location history (SLH), e.g., shopping malls → restaurants → cinemas. Then, we measure the similarity between different users' SLHs by using our maximal travel match (MTM) algorithm. The advantage of our approach lies in two aspects. First, SLH carries more semantic meanings of a user's interests beyond low-level geographic positions. Second, our approach can estimate the similarity between two users without overlaps in the geographic spaces, e.g., people living in different cities. We evaluate our method based on a real-world GPS dataset collected by 109 users in a period of 1 year. As a result, SLH-MTM outperforms the related works [4].

GPS-equipped taxis can be regarded as mobile sensors probing traffic flows on road surfaces, and taxi drivers are usually experienced in finding the fastest (quickest) route to a destination based on their knowledge. In this paper, we mine smart driving directions from the historical GPS trajectories of a large number of taxis, and provide a user with the practically fastest route to a given destination at a given departure time. In our approach, we propose a time-dependent landmark graph, where a node (landmark) is a road segment frequently traversed by taxis, to model the intelligence of taxi drivers and the properties of dynamic road networks. Then, a Variance-Entropy-Based Clustering approach is devised to estimate the distribution of travel time between two landmarks in different time slots. Based on this graph, we design a two-stage routing algorithm to compute the practically fastest route. We build our system based on a real-world trajectory dataset generated by over 33,000 taxis in a period of 3 months, and evaluate the system by conducting both synthetic experiments and in-the-field evaluations. As a result, 60-70% of the routes suggested by our method are faster than the competing methods, and 20% of the routes share the same results. On average, 50% of our routes are at least 20% faster than the competing approaches.

The quality of a local search engine, such as Google and Bing Maps, heavily relies on its geographic datasets. Typically, these datasets are obtained from multiple sources, e.g., different vendors or public yellow-page websites. Therefore, the same location entity, like a restaurant, might have multiple records with slightly different presentations of title and address in different data sources. For instance, 'Seattle Premium Outlets' and 'Seattle Premier Outlet Mall' describe the same Outlet located in the same place while their titles are not identical. This will cause many nearly-duplicated records in a location database, which would bring trouble to data management and make users confused by the various search results of a query. To detect these nearly duplicated records, we propose a machine-learning-based approach, which is comprised of three steps: candidate selection, feature extraction and training/inference. Three key features consisting of name similarity, address similarity and category similarity, as well as corresponding metrics, are proposed to model the differences between two entity records. We evaluate our method with intensive experiments based on a large-scale real dataset. As a result, both the precision and recall of our method exceeded 90%.

The emerging location-detection devices together with ubiquitous connectivity have enabled a large variety of location-based services (LBS). Unfortunately, LBS may threaten the users' privacy. K-anonymity cloaking the user location to K-anonymizing spatial region (K-ASR) has been extensively studied to protect privacy in LBS. Traditional K-anonymity method needs complex query processing algorithms at the server side. SpaceTwist rectifies the above shortcoming of traditional K-anonymity since it only requires incremental nearest neighbor (INN) queries processing techniques at the server side. However, Space Twist may fail since it cannot guarantee K-anonymity. In this paper, our proposed framework, called KAWCR (K-anonymity Without Cloaked Region), rectifies the shortcomings and retains the advantages of the above two techniques. KAWCR only needs the server to process INN queries and can guarantee that the users issuing the query is indistinguishable from at least K-1 other users. The extensive experimental results show that the communication cost of KAWCR for kNN queries is lower than that of both traditional K-anonymity and SpaceTwist.

Matching a raw GPS trajectory to roads on a digital map is often referred to as the Map Matching problem. However, the occurrence of the low-sampling-rate trajectories (e.g. one point per 2 minutes) has brought lots of challenges to existing map matching algorithms. To address this problem, we propose an Interactive Voting-based Map Matching (IVMM) algorithm based on the following three insights: 1) The position context of a GPS point as well as the topological information of road networks, 2) the mutual influence between GPS points (i.e., the matching result of a point references the positions of its neighbors; in turn, when matching its neighbors, the position of this point will also be referenced), and 3) the strength of the mutual influence weighted by the distance between GPS points (i.e., the farther distance is the weaker influence exists). In this approach, we do not only consider the spatial and temporal information of a GPS trajectory but also devise a voting-based strategy to model the weighted mutual influences between GPS points. We evaluate our IVMM algorithm based on a user labeled real trajectory dataset. As a result, the IVMM algorithm outperforms the related method (ST-Matching algorithm).

People's location histories imply the location correlation that states the relations between geographical locations in the space of human behavior. With the correlation, we can enable many valuable services, such as location recommendation and sales promotion. In this paper, by taking into account a user's travel experience (knowledge) and the sequentiality that locations have been visited, we learn the location correlation from a large number of user-generated GPS trajectories. Using the location correlation, we conduct a personalized location recommendation system, which is evaluated based on a real-world GPS dataset collected by 112 users over a period of 1.5 years. As a result, our method outperforms that using the Pearson correlation.

Photo sharing is one of the most popular Web services. Photo sharing sites provide functions to add tags and geo-tags to photos to make photo organization easy. Considering that people take photos to record something that attracts them, geo-tagged photos are a rich data source that reflects people's memorable events associated with locations. In this paper, we focus on geo-tagged photos and propose a method to detect people's frequent trip patterns, i.e., typical sequences of visited cities and durations of stay as well as descriptive tags that characterize the trip patterns. Our method first segments photo collections into trips and categorizes them based on their trip themes, such as visiting landmarks or communing with nature. Our method mines frequent trip patterns for each trip theme category. We crawled 5.7 million geo-tagged photos and performed photo trip pattern mining. The experimental result shows that our method outperforms other baseline methods and can correctly segment photo collections into photo trips with an accuracy of 78%. For trip categorization, our method can categorize about 80% of trips using tags and titles of photos and visited cities as features. Finally, we illustrate interesting examples of trip patterns detected from our dataset and show an application with which users can search frequent trip patterns by querying a destination, visit duration, and trip theme on the trip.

Trajectory search has long been an attractive and challenging topic which blooms various interesting applications in spatial-temporal databases. In this work, we study a new problem of searching trajectories by locations, in which context the query is only a small set of locations with or without an order specified, while the target is to find the k Best-Connected Trajectories (k-BCT) from a database such that the k-BCT best connect the designated locations geographically. Different from the conventional trajectory search that looks for similar trajectories w.r.t. shape or other criteria by using a sample query trajectory, we focus on the goodness of connection provided by a trajectory to the specified query locations. This new query can benefit users in many novel applications such as trip planning.

With the increasing popularity of location-based services, such as tour guide and location-based social network, we now have accumulated many location data on the Web. In this paper, we show that, by using the location data based on GPS and users' comments at various locations, we can discover interesting locations and possible activities that can be performed there for recommendations. Our research is highlighted in the following location-related queries in our daily life: 1) if we want to do something such as sightseeing or food-hunting in a large city such as Beijing, where should we go? 2) If we have already visited some places such as the Bird's Nest building in Beijing's Olympic park, what else can we do there? By using our system, for the first question, we can recommend her to visit a list of interesting locations such as Tiananmen Square, Bird's Nest, etc. For the second question, if the user visits Bird's Nest, we can recommend her to not only do sightseeing but also to experience its outdoor exercise facilities or try some nice food nearby. To achieve this goal, we first model the users' location and activity histories that we take as input. We then mine knowledge, such as the location features and activity-activity correlations from the geographical databases and the Web, to gather additional inputs. Finally, we apply a collective matrix factorization method to mine interesting locations and activities, and use them to recommend to the users where they can visit if they want to perform some specific activities and what they can do if they visit some specific places. We empirically evaluated our system using a large GPS dataset collected by 162 users over a period of 2.5 years in the real-world. We extensively evaluated our system and showed that our system can outperform several state-of-the-art baselines.

Map search engines, such as Google Maps, Yahoo! Maps, and Microsoft Live Maps, allow users to explicitly specify a target geographic location, either in keywords or on the map, and to search businesses, people, and other information of that location. In this article, we report a first study on a million-entry map search log. We identify three key attributes of a map search record—the keyword query, the target location and the user location, and examine the characteristics of these three dimensions separately as well as the associations between them. Comparing our results with those previously reported on logs of general search engines and mobile search engines, including those for geographic queries, we discover the following unique features of map search: (1) People use longer queries and modify queries more frequently in a session than in general search and mobile search; People view fewer result pages per query than in general search; (2) The popular query topics in map search are different from those in general search and mobile search; (3) The target locations in a session change within 50 kilometers for almost 80% of the sessions; (4) Queries, search target locations and user locations (both at the city level) all follow the power law distribution; (5) One third of queries are issued for target locations within 50 kilometers from the user locations; (6) The distribution of a query over target locations appears to follow the geographic location of the queried entity.

User mobility has given rise to a variety of Web applications, in which the global positioning system (GPS) plays many important roles in bridging between these applications and end users. As a kind of human behavior, transportation modes, such as walking and driving, can provide pervasive computing systems with more contextual information and enrich a user's mobility with informative knowledge. In this article, we report on an approach based on supervised learning to automatically infer users' transportation modes, including driving, walking, taking a bus and riding a bike, from raw GPS logs. Our approach consists of three parts: a change point-based segmentation method, an inference model and a graph-based post-processing algorithm. First, we propose a change point-based segmentation method to partition each GPS trajectory into separate segments of different transportation modes. Second, from each segment, we identify a set of sophisticated features, which are not affected by differing traffic conditions (e.g., a person's direction when in a car is constrained more by the road than any change in traffic conditions). Later, these features are fed to a generative inference model to classify the segments of different modes. Third, we conduct graph-based postprocessing to further improve the inference performance. This postprocessing algorithm considers both the commonsense constraints of the real world and typical user behaviors based on locations in a probabilistic manner. The advantages of our method over the related works include three aspects. (1) Our approach can effectively segment trajectories containing multiple transportation modes. (2) Our work mined the location constraints from user-generated GPS logs, while being independent of additional sensor data and map information like road networks and bus stops. (3) The model learned from the dataset of some users can be applied to infer GPS data from others. Using the GPS logs collected by 65 people over a period of 10 months, we evaluated our approach via a set of experiments. As a result, based on the change-point-based segmentation method and Decision Tree-based inference model, we achieved prediction accuracy greater than 71 percent. Further, using the graph-based post-processing algorithm, the performance attained a 4-percent enhancement.

Scalable image retrieval systems usually involve hierarchical quantization of local image descriptors, which produces a visual vocabulary for inverted indexing of images. Although hierarchical quantization has the merit of retrieval efficiency, the resulting visual vocabulary representation usually faces two crucial problems: (1) hierarchical quantization errors and biases in the generation of “visual words”; (2) the model cannot adapt to database variance. In this paper, we describe an unsupervised optimization strategy in generating the hierarchy structure of visual vocabulary, which produces a more effective and adaptive retrieval model for large-scale search. We adopt a novel Density-based Metric Learning (DML) algorithm, which corrects word quantization bias without supervision in hierarchy optimization, based on which we present a hierarchical rejection chain for efficient online search based on the vocabulary hierarchy. We also discovered that by hierarchy optimization, efficient and effective transfer of a retrieval model across different databases is feasible. We deployed a large-scale image retrieval system using a vocabulary tree model to validate our advances. Experiments on UKBench and street-side urban scene databases demonstrated the effectiveness of our hierarchy optimization approach in comparison with state-of-the-art methods.

Map-matching is the process of aligning a sequence of observed user positions with the road network on a digital map. It is a fundamental pre-processing step for many applications, such as moving object management, traffic flow analysis, and driving directions. In practice there exists huge amount of low-sampling-rate (e.g., one point every 2--5 minutes) GPS trajectories. Unfortunately, most current map-matching approaches only deal with high-sampling-rate (typically one point every 10--30s) GPS data, and become less effective for low-sampling-rate points as the uncertainty in data increases. In this paper, we propose a novel global map-matching algorithm called ST-Matching for low-sampling-rate GPS trajectories. ST-Matching considers (1) the spatial geometric and topological structures of the road network and (2) the temporal/speed constraints of the trajectories. Based on spatio-temporal analysis, a candidate graph is constructed from which the best matching path sequence is identified. We compare ST-Matching with the incremental algorithm and Average-Fréchet-Distance (AFD) based global map-matching algorithm. The experiments are performed both on synthetic and real dataset. The results show that our ST-matching algorithm significantly outperform incremental algorithm in terms of matching accuracy for low-sampling trajectories. Meanwhile, when compared with AFD-based global algorithm, ST-Matching also improves accuracy as well as running time.

The advance of location-acquisition technologies enables people to record their location histories with spatio-temporal datasets, which imply the correlation between geographical regions. This correlation indicates the relationship between locations in the space of human behavior, and can enable many valuable services, such as sales promotion and location recommendation. In this paper, by taking into account a user's travel experience and the sequentiality locations have been visited, we propose an approach to mine the correlation between locations from a large number of users' location histories. We conducted a personalized location recommendation system using the location correlation, and evaluated this system with a large-scale real-world GPS dataset. As a result, our method outperforms the related work using the Pearson correlation.

A critical component of today's commercial search engines is an advertisement platform. The current state-of-the-art of such platforms is primarily based on advanced keyword matching to determine the relevance of advertisements for users' queries. However, such keyword matching techniques suffer from missing user intent when the query domain is visual as opposed to textual. To handle such a domain, we propose a new advertisement platform which allows search engine advertisers to bid on images instead of just plain text. The main components of this platform include an advertisement editorial tool, ROI detection, image content understanding, and image matching modules. This platform is suitable for application scenarios where images are the main input or consumed content, for example, in Multimedia Messaging Service (MMS) or content based image retrieval. We demonstrated the effectiveness of our proposed advertisement platform solution when used in a mobile visual search scenario involving querying for real world billboards.

In this demonstration, we present VisiAds, a vision-based online advertising platform which allows advertisers to bid on image content as opposed to text keywords. The demonstration shows an implementation of the platform designed specifically for mobile application scenarios. It can be used by advertisers to trigger and deliver multiple advertisements based on a query image captured from a consumer-grade cellular phone with an embedded camera.

The increasing pervasiveness of location-acquisition technologies (GPS, GSM networks, etc.) enables people to conveniently log their location history into spatial-temporal data, thus giving rise to the necessity as well as opportunity to discovery valuable knowledge from this type of data. In this paper, we propose the novel notion of individual life pattern, which captures individual's general life style and regularity. Concretely, we propose the life pattern normal form (the LP-normal form) to formally describe which kind of life regularity can be discovered from location history; then we propose the LP-Mine framework to effectively retrieve life patterns from raw individual GPS data. Our definition of life pattern focuses on significant places of individual life and considers diverse properties to combine the significant places. LP-Mine is comprised of two phases: the modelling phase and the mining phase. The modelling phase pre-processes GPS data into an available format as the input of the mining phase. The mining phase applies separate strategies to discover different types of pattern. Finally, we conduct extensive experiments using GPS data collected by volunteers in the real world to verify the effectiveness of the framework.

GeoLife2.0 is a GPS-data-driven social networking service where people can share life experiences and connect to each other with their location histories. By mining peoplepsilas location history, GeoLife can measure the similarity between users and perform personalized friend recommendation for an individual. Later, we can predict the individualpsilas interest level in the locations visited by their friends while have not been found by them. The locations with relatively high interesting level can be recommended. Therefore, GeoLife2.0 can expand a userpsilas social network, provide them with a trustworthy resource matching their interests and help them sponsor geo-related activities like cycling with minimal effort.

Recent years have witnessed great prosperity in community-contributed multimedia. Discovering, extracting, and summarizing knowledge from these data enables us to make better sense of the world. In this paper, we report our work on mining famous city landmarks from blogs for personalized tourist suggestions. Our main contribution is a graph modeling framework to discover city landmarks by mining blog photo correlations with community supervision. This modeling fuses context, content, and community information in a style that simulates both static (PageRank) and dynamic (HITS) ranking models to highlight representative data from the consensus of blog users.

This paper introduces the architecture of our location sensitive indexing model which is used in a platform designed to deliver advertisements to users who primarily utilize images as queries instead of textual keywords. The indexing model facilitates an advertiser's ability to bid on images, such as billboards or logos, in order to obtain user feedback in judging image attractiveness. Additionally, the model enables automatic evaluation of advertisement popularity by mining users' query logs, which is critical for generating advertisement recommendations. The location sensitive architecture of this model enables effective and efficient functionality in large-scale scenarios. In the model's structure, our Location Sensitive Visual Indexing model (LSVI) incorporates location information that subdivides geographical regions for precise and localized image matching. By collecting feedback from mobile users, location-based mining can also help discover popular advertisements as well as their representative images. We have deployed our platform into a real-world advertising system in Beijing, China, which demonstrates effective results in comparative studies with both alternative and state-of-the-art approaches.

Geographical context is very important for images. Millions of images on the Web have been already assigned latitude and longitude information. Due to the rapid proliferation of such images with geographical context, it is still difficult to effectively search and browse them, since we do not have ways to decide their relevance. In this paper, we focus on the geographical relevance of images, which is defined as to what extent the main objects in an image match landmarks at the location where the image was taken. Recently, researchers have proposed to use game based approaches to label large scale data such as Web images. However, previous works have not shown the quality of collected game logs in detail and how the logs can improve existing applications. To answer these questions, we design and implement a Web-based and multi-player game to collect human knowledge while people are enjoying the game. Then we thoroughly analyze the game logs obtained during a three week study with 147 participants and propose methods to determine the image geographical relevance. In addition, we conduct an experiment to compare our methods with a commercial search engine. Experimental results show that our methods dramatically improve image search relevance. Furthermore, we show that we can derive geographically relevant objects and their salient portion in images, which is valuable for a number of applications such as image location recognition.

The increasing availability of GPS-enabled devices is changing the way people interact with the Web, and brings us a large amount of GPS trajectories representing people's location histories. In this paper, based on multiple users' GPS trajectories, we aim to mine interesting locations and classical travel sequences in a given geospatial region. Here, interesting locations mean the culturally important places, such as Tiananmen Square in Beijing, and frequented public areas, like shopping malls and restaurants, etc. Such information can help users understand surrounding locations, and would enable travel recommendation. In this work, we first model multiple individuals' location histories with a tree-based hierarchical graph (TBHG). Second, based on the TBHG, we propose a HITS (Hypertext Induced Topic Search)-based inference model, which regards an individual's access on a location as a directed link from the user to that location. This model infers the interest of a location by taking into account the following three factors. 1) The interest of a location depends on not only the number of users visiting this location but also these users' travel experiences. 2) Users' travel experiences and location interests have a mutual reinforcement relationship. 3) The interest of a location and the travel experience of a user are relative values and are region-related. Third, we mine the classical travel sequences among locations considering the interests of these locations and users' travel experiences. We evaluated our system using a large GPS dataset collected by 107 users over a period of one year in the real world. As a result, our HITS-based inference model outperformed baseline approaches like rank-by-count and rank-by-frequency. Meanwhile, when considering the users' travel experiences and location interests, we achieved a better performance beyond baselines, such as rank-by-count and rank-by-interest, etc.

This paper presents an efficient and effective solution for retrieving image near-duplicate (IND) from image database. We introduce the coherent phrase model which incorporates the coherency of local regions to reduce the quantization error of the bag-of-words (BoW) model. In this model, local regions are characterized byvisual phraseof multiple descriptors instead of visual word of single descriptor. We propose two types of visual phrase to encode the coherency in feature and spatial domain, respectively. The proposed model reduces the number of false matches by using this coherency and generates sparse representations of images. Compared to other method, the local coherencies among multiple descriptors of every region improve the performance and preserve the efficiency for IND retrieval. The proposed method is evaluated on several benchmark datasets for IND retrieval. Compared to the state-of-the-art methods, our proposed model has been shown to significantly improve the accuracy of IND retrieval while maintaining the efficiency of the standard bag-of-words model. The proposed method can be integrated with other extensions of BoW.

We propose a new Web page transformation method to facilitate Web browsing on handheld devices such as Personal Digital Assistants (PDAs). In our approach, an original Web page that does not fit on the screen is transformed into a set of subpages, each of which fits on the screen. This transformation is done through slicing the original page into page blocks iteratively, with several factors considered. These factors include the size of the screen, the size of each page block, the number of blocks in each transformed page, the depth of the tree hierarchy that the transformed pages form, as well as the semantic coherence between blocks. We call the tree hierarchy of the transformed pages an SP-tree. In an SP-tree, an internal node consists of a textually enhanced thumbnail image with hyperlinks, and a leaf node is a block extracted from a subpage of the original Web page. We adaptively adjust the fanout and the height of the SP-tree so that each thumbnail image is clear enough for users to read, while at the same time, the number of clicks needed to reach a leaf page is few. Through this transformation algorithm, we preserve the contextual information in the original Web page and reduce scrolling. We have implemented this transformation module on a proxy server and have conducted usability studies on its performance. Our system achieved a shorter task completion time compared with that of transformations from the Opera browser in nine of ten tasks. The average improvement on familiar pages was 44%. The average improvement on unfamiliar pages was 37%. Subjective responses were positive.

The pervasiveness of location-acquisition technologies (GPS, GSM networks, etc.) enable people to conveniently log the location histories they visited with spatio-temporal data. The increasing availability of large amounts of spatio-temporal data pertaining to an individual's trajectories has given rise to a variety of geographic information systems, and also brings us opportunities and challenges to automatically discover valuable knowledge from these trajectories. In this paper, we move towards this direction and aim to geographically mine the similarity between users based on their location histories. Such user similarity is significant to individuals, communities and businesses by helping them effectively retrieve the information with high relevance. A framework, referred to as hierarchical-graph-based similarity measurement (HGSM), is proposed for geographic information systems to consistently model each individual's location history and effectively measure the similarity among users. In this framework, we take into account both the sequence property of people's movement behaviors and the hierarchy property of geographic spaces. We evaluate this framework using the GPS data collected by 65 volunteers over a period of 6 months in the real world. As a result, HGSM outperforms related similarity measures, such as the cosine similarity and Pearson similarity measures.

Co-location pattern discovery is to find classes of spatial objects that are frequently located together. For example, if two categories of businesses often locate together, they might be identified as a co-location pattern; if several biologic species frequently live in nearby places, they might be a co-location pattern. Most existing co-location pattern discovery methods are generate-and-test methods, that is, generate candidates, and test each candidate to determine whether it is a co-location pattern. In the test step, we identify instances of a candidate to obtain its prevalence. In general, instance identification is very costly. In order to reduce the computational cost of identifying instances, we propose a density based approach. We divide objects into partitions and identifying instances in dense partitions first. A dynamic upper bound of the prevalence for a candidate is maintained. If the current upper bound becomes less than a threshold, we stop identifying its instances in the remaining partitions. We prove that our approach is complete and correct in finding co-location patterns. Experimental results on real data sets show that our method outperforms a traditional approach.

Both recognizing human behavior and understanding a user's mobility from sensor data are critical issues in ubiquitous computing systems. As a kind of user behavior, the transportation modes, such as walking, driving, etc., that a user takes, can enrich the user's mobility with informative knowledge and provide pervasive computing systems with more context information. In this paper, we propose an approach based on supervised learning to infer people's motion modes from their GPS logs. The contribution of this work lies in the following two aspects. On one hand, we identify a set of sophisticated features, which are more robust to traffic condition than those other researchers ever used. On the other hand, we propose a graph-based post-processing algorithm to further improve the inference performance. This algorithm considers both the commonsense constraint of real world and typical user behavior based on location in a probabilistic manner. Using the GPS logs collected by 65 people over a period of 10 months, we evaluated our approach via a set of experiments. As a result, based on the change point-based segmentation method and Decision Tree-based inference model, the new features brought an eight percent improvement in inference accuracy over previous result, and the graph-based post-processing achieve a further four percent enhancement.

Recently, there has been growing interest in mining co-location visual patterns from a collection of images. To find a proper usage of visual patterns in near-duplicate image retrieval systems, we study a TF-IDF weighting function for visual patterns. We show usage of TF and IDF respectively in this weighting function. Experiments demonstrate that 1) visual patterns and words should be weighted separately; 2) visual patterns consist of more frequent visual words would be more important; 3) visual patterns should be given lower weight values than visual words in near-duplicate image retrieval tasks.

This work aims at developing a scalable vision-based location recognition system where the backend database can be updated incrementally. Our proposed framework enables incremental indexing of vocabulary tree model, which efficiently includes new data into model refinement without re-generating entire model from overall dataset. An adaption trigger criterion is presented to lessen system computational cost, which is achieved by density-based relative entropy estimation between original dataset and newly coming data. Experiments on Seattle urban scene datasets with over 20K street-side images show the effectiveness of our work.

This work introduces a novel data mining scheme, spatial pyramid mining, to discover association rules at multiple resolutions in order to identify frequent spatial configurations of local features that correspond to classes of logos appearing in real world scenes. By indexing representative examples by the mined rules we can efficiently detect a variety of different lettering or design marks associated with a brand. Features in an image are marked by matching rules to representative examples selected via a weighted cosine similarity measure. Logos are localized in an image via density-based clustering of matched features. Precision vs. recall curves are presented for experiments on a dataset of web images of nearly 1,000 images containing seven popular logo types.

The increasing popularity of GPS device has boosted many Web applications where people can upload, browse and exchange their GPS tracks. In these applications, spatial or temporal search function could provide an effective way for users to retrieve specific GPS tracks they are interested in. However, existing spatial-temporal index for trajectory data has not exploited the characteristic of user behavior in these online GPS track sharing applications. In most cases, when sharing a GPS track, people are more likely to upload GPS data of the near past than the distant past. Thus, the interval between the end time of a GPS track and the time it is uploaded, if viewed as a random variable, has a skewed distribution. In this paper, we first propose a probabilistic model to simulate user behavior of uploading GPS tracks onto an online sharing application. Then we propose a flexible spatio-temporal index scheme, referred to as Compressed Start-End Tree (CSE-tree), for large-scale GPS track retrieval. The CSE-tree combines the advantages of B+ Tree and dynamic array, and maintains different index structure for data with different update frequency. Experiments using synthetic data show that CSE-tree outperforms other schemes in requiring less index size and less update cost while keeping satisfactory retrieval performance.

The increasing popularity of GPS device has boosted many applications where more and more GPS logs have been accumulating continuously. Managing and understanding the collected GPS data are two important issues for these applications. On one hand, by indexing the increasing GPS data, we can provide effective retrieval method for users to find the corresponding GPS data interests them. On the other hand, by understanding user's GPS data, we are more likely to enable novel services which would stimulate people's passion on contributing GPS data in turn. However, so far, GPS data are still used directly without much understanding. In our project, referred to as GeoLife, we focus on visualization, organization, fast retrieval, and effective understanding of GPS track logs for both personal and public use. It not only provides a powerful platform for people to effectively manage their GPS data but also help them well understand a person's past experience from GPS data.

Geographic information has spawned many novel Web applications where global positioning system (GPS) plays important roles in bridging the applications and end users. Learning knowledge from users ’ raw GPS data can provide rich context information for both geographic and mobile applications. However, so far, raw GPS data are still used directly without much understanding. In this paper, an approach based on supervised learning is proposed to automatically infer transportation mode from raw GPS data. The transportation mode, such as walking, driving, etc., implied in a user’s GPS data can provide us valuable knowledge to understand the user. It also enables context-aware computing based on user’s present transportation mode and design of an innovative user interface for Web users. Our approach consists of three parts: a change point-

The popularity of mobile devices, such as PDAs and SmartPhones, has grown rapidly over the last couple of years. Though most users still perform searches using desktop computers, it is expected that more and more people will also search the Web while they are on the move. In addition to text-based keyword queries, mobile devices can support richer and hybrid queries such as images, audio, video, and their combinations. In this paper, we will discuss mobile search systems that support image queries and audio queries, covering typical designs for mobile visual and audio search, as well as the opportunities and challenges. Specifically, we will present an in-depth study of two real systems we have developed: product image categorization and mobile ringtone search, which use image queries and audio queries, respectively. Experimental results on real-life data demonstrate their effectiveness and efficiency.

In recent years, some computer vision algorithms such as SIFT (Scale Invariant Feature Transform) have been employed in image similarity match to perform image-based search applications. However, with the increasing scale of image databases, centralized image retrieval system no longer provide adequate prompt search. In this paper, we design a scalable distributed architecture, which is analog to web search engine, for efficient large-scale image retrieval. In our distributed architecture, images are partitioned to multiple servers and an index is built. Administrated by a controlling server, each distributed server matches query image in its own image sub-collection in parallel and returns the intermediate search results, a list of images similar to query image, to the controlling server for further re-ranking and merging. An evaluation of the results shows that our distributed architecture removes the limitation of a centralized image retrieval system. By performing reasonable indexing, merging and ranking strategies, the precision level of search is near to that performed on stand-alone retrieval systems indexing all images.

In this paper, we study the problem of learning block classification models to estimate block functions. We distinguish general models, which are learned across multiple sites, and site-specific models, which are learned within individual sites. We further consider several factors that affect the learning process and model effectiveness. These factors include the layout features, the content features, the classifiers, and the term selection methods. We have empirically evaluated the performance of the models when the factors are varied. Our main results are that layout features do better than content features for learning both general and site-specific models.

In this paper, we presented an image search service for mobile users. It can be used to acquire related information by taking and sending pictures to the server, for example, getting book reviews by a photo of the cover. The key problem here is to find images that contain the same prominent object as that in the query image. In the literature, local feature based image matching has been proven to outperform those based on global features. When using local features, however, one query image may contain thousands of high dimensional feature vectors. Each feature vector needs to match against millions of features in the database. Therefore, it is critical to design an efficient search scheme. Our proposed matching approach was based on identifying semi-local visual parts from multiple query images. Experiments on two real-world datasets showed that this approach was superior to conventional solutions

In image browsing, the resolution greatly affects user’s experience. If an image is down-scaled too much, a considerable amount of information within it will be lost. In this paper, we studied the problem of "What is a sufficient display resolution or scale for an image or an image region?" This problem arises in many real-life applications including image browsing on mobile devices, image adaptation and progressive image delivery. Kullback-Leibler (K-L) distance is employed to measure the information loss and the sufficient display scale is selected based on the information loss curve during image down-sampling. Since the images are presented to viewers finally, some visual characteristics are also taken into account to ensure the precision of the measurement. A user study was carried out to evaluate the performance of our approach. Experimental results show that the approach is in good accord with human perception.

With the pervasive use of camera phones, the embedded camera has been considered as a promising HCI manner for mobiles. With necessary technologies, it is possible to become a powerful tool to acquire the information in daily life. We have designed and implemented a system named Photo-to-Search to carry out queries from camera phones simply by taking some photos of interested objects. The captured pictures are compared with a large amount of Web images to select the ones which contain the same prominent object. Consequently, the related information is extracted from the Web pages where the matched images locate. In our demo, data of large buildings, storefronts and products are collected and these kinds of queries are specifically demonstrated to show the efficiency and the effectiveness of our system.

Currently, people gain easy access to an increasingly diverse range of mobile devices such as personal digital assistants (PDAs), smart phones, and handheld computers. As dynamic content has become dominant on the fast-growing World Wide Web (C. Yuan et al., 2003), it is necessary to provide effective ways for the users to access such prevalent Web content in a mobile computing environment. During a course of browsing dynamic content on mobile devices, the requested content is first dynamically generated by remote Web server, then transmitted over a wireless network, and, finally, adapted for display' on small screens. This leads to considerable latency and processing load on mobile devices. By integrating a novel Web content adaptation algorithm and an enhanced caching strategy, we propose an adaptive scheme called MobiDNA for serving dynamic content in a mobile computing environment. To validate the feasibility and effectiveness of the proposed MobiDNA system, we construct an experimental testbed to investigate its performance. Experimental results demonstrate that this scheme can effectively improve mobile dynamic content browsing, by improving Web content readability on small displays, decreasing mobile browsing latency, and reducing wireless bandwidth consumption

Pictures have become increasingly common and popular in mobile communications. However, due to the limitation of mobile devices, there is a need to develop new technologies to facilitate the browsing of large pictures on the small screen. In this paper, we propose a set of novel approaches which are able to aid or automate common image browsing tasks on mobile devices. All of these approaches are based on an image attention model which is employed to illustrate the information structure within an image. An efficient algorithm to generate the optimal browsing path based on the information foraging theory is presented. Experimental evaluations of the proposed mechanism indicate that our approach is an effective way for viewing large images on small displays

Mobile devices which can capture and view pictures are becoming increasingly common in our life. The limitation of these small-form-factor devices makes the user experience of image browsing quite different from that on desktop PCs. In this paper, we first present a user study on how users interact with a mobile image browser with basic functions. We found that on small displays, users tend to use more zooming and scrolling actions in order to view interesting regions in detail. From this fact, we designed a new method to detect user interest maps and extract user attention objects from the image browsing log. This approach is more efficient than image-analysis based methods and can better represent users' actual interest. A smart image viewer was then developed based on user interest analysis. A second experiment was carried out to study how users behave with such a viewer. Experimental results demonstrate that the new smart features can improve the browsing efficiency and are a good compliment to traditional image browsers.

There is more and more commercial and research interest in location-based web search, i.e. finding web content whose topic is related to a particular place or region. In this type of search, location information should be indexed as well as text information. However, the index of conventional text search engine is set-oriented, while location information is two-dimensional and in Euclidean space. This brings new research problems on how to efficiently represent the location attributes of web pages and how to combine two types of indexes. In this paper, we propose to use a hybrid index structure, which integrates inverted files and R*-trees, to handle both textual and location aware queries. Three different combining schemes are studied: (1) inverted file and R*-tree double index, (2) first inverted file then R*-tree, (3) first R*-tree then inverted file. To validate the performance of proposed index structures, we design and implement a complete location-based web search engine which mainly consists of four parts: (1) an extractor which detects geographical scopes of web pages and represents geographical scopes as multiple MBRs based on geographical coordinates; (2) an indexer which builds hybrid index structures to integrate text and location information; (3) a ranker which ranks results by geographical relevance as well as non-geographical relevance; (4) an interface which is friendly for users to input location-based search queries and to obtain geographical and textual relevant results. Experiments on large real-world web dataset show that both the second and the third structures are superior in query time and the second is slightly better than the third. Additionally, indexes based on R*-trees are proven to be more efficient than indexes based on grid structures.

Currently, a crucial challenge is raised on how to manage a large amount of images on the Web. Due to a real synergy between an image and its location, we propose an automatic solution to annotate contextual location information for WWW images. We construct an image importance model to acquire the dominant images in a page that comprise contextual surrounding text. For each acquired image, we develop an effective algorithm to compute location from its contextual text. We apply our approach to 1,000 pages from various Websites for image location annotation. The experiments demonstrated that more than 30% WWW images are related with geographic location information, and our solution can achieve the satisfactory results. Finally, we present some potential applications involving the utilization of image location information

This paper proposes an ambient interactive and responsive environment (AIRE) to improve user's image browsing experiences on the small-form-factor devices. This solution is characterized as two aspects: (1) establishing an ambient interactive communication across various devices; and (2) designing a distributed interface that crosses various devices to overcome the display constraint in mobile devices. In the AIRE system, a two-level image browsing scheme is designed to meet users' various image browsing needs.

Digital photos have become increasingly common and popular in mobile communications. However, due to the distribution of these photos captured in various devices, there is a need to develop new technologies to facilitate the sharing of large image collections across these devices for users. In this paper, we propose A Synchronous Approach for Photo sharing across multiple devices (ASAP). The ASAP provides a hierarchical two-level synchronization scheme, namely image-level and region-level. In the ASAP, a user’s interaction with any device automatically leads to a series of synchronous updates in other devices. Thus, the ASAP simultaneously presents similar images across devices in a way that allows automatic synchronization of images based on user interactions. Experimental evaluations indicate that it is effective and useful to improve image browsing experience across devices for users.

It is expected that more and more people would search the Web when they are on the move. Though conventional search engines can be directly visited from mobile devices with Web browsing capabilities, the information is not as conveniently accessible from a handheld device as it is from desktops. Existing information discovery mechanisms for searching the Web are not well-suited to mobile devices. In this paper, a block importance model is employed to assign importance values to different segments of a Web page, in order to extract and present more condensed search results to mobile users. Based on the block importance model, three presentations for displaying the result pages in different levels of detail have been designed to reduce both the number of user interactions and the overall search time. A set of user study experiments have been carried out to compare the three presentations and a commercial service on typical mobile devices. Experimental results show that our approaches can help users to explore Web search results more efficiently.