Energy efficiency has become a critical requirement, driving recommendation systems for resource-constrained environments such as edge devices. Model quantization offers an effective way to build low-bitwidth models while preserving accuracy. However, user–item interaction graphs contain numerous nodes and complex topological structures, leading nodes to exhibit unique similarities and differences. Existing quantization methods uniformly process parameters in high-dimensional DNN layers (e.g., linear, convolutional, or attention layers), while inadequately capturing such similarities among node embeddings. This paper proposes GraphQ, a chunk-wise quantization framework for graph collaborative filtering that supports both the training and post-training phases in a unified perspective. Our core idea is to adaptively partition node embeddings into multiple chunks based on the distribution of embedding values, and then apply chunk-wise quantization. Specifically, for quantization-aware training (QAT), we introduce learnable low-precision quantization factors that partition node embeddings into multiple chunks and are dynamically updated following message passing. For post-training quantization (PTQ), we first cluster nodes and then partition their dimensions into chunks for weight clipping. Extensive experiments on four real-world datasets show that GraphQ outperforms state-of-the-art QAT methods by an average of 27.49% in Recall@10 under the 256-dimensional embedding and 2-bit settings, and surpasses PTQ methods by 78.64% on average under 4-bit settings.

Heterophily is a prevalent property of real-world graphs and is well known to impair the performance of homophilic Graph Neural Networks (GNNs). Prior work has attempted to adapt GNNs to heterophilic graphs through non-local neighbor extension or architecture refinement. However, the fundamental reasons behind misclassifications remain poorly understood. In this work, we take a novel perspective by examining recurring inductive subgraphs, empirically and theoretically showing that they act as spurious shortcuts that mislead GNNs and reinforce non-causal correlations in heterophilic graphs. To address this, we adopt a causal inference perspective to analyze and correct the biased learning behavior induced by shortcut inductive subgraphs. We propose a debiased causal graph that explicitly blocks confounding and spillover paths responsible for these shortcuts. Guided by this causal graph, we introduce Causal Disentangled GNN (CD-GNN), a principled framework that disentangles spurious inductive subgraphs from true causal subgraphs by explicitly blocking non-causal paths. By focusing on genuine causal signals, CD-GNN substantially improves the robustness and accuracy of node classification in heterophilic graphs. Extensive experiments on real-world datasets not only validate our theoretical findings but also demonstrate that our proposed CD-GNN outperforms state-of-the-art heterophily-aware baselines.

Session-based recommender systems face significant challenges in accurately predicting user preferences due to the limited availability of long-term historical interactions. While recent advances in deep learning and graph-based approaches have improved recommendation performance, the temporal aspects of user interactions remain underutilized. This paper identifies three critical temporal challenges in session-based recommendations: interest shifts indicated by long intervals between interactions, interaction noise from brief engagements, and system popularity effects during high-traffic periods. To address these challenges, we propose a novel Dual-channel Time-aware Graph Attention Network (DT-GAT) to incorporate temporal signal, i.e., time intervals between interactions and time differences between sessions, into session representations from both item and session perspectives. The item-wise learning channel employs a temporal graph attention network to capture interest shifts and filter interaction noise, while the session-wise learning channel utilizes a temporal graph attention network to handle inconsistent popularity trends. Additionally, we introduce a multi-temporal window processing mechanism to construct robust session representations that effectively capture short-term interests while filtering noise. Extensive experiments conducted on three real-world datasets demonstrate that DT-GAT consistently outperforms state-of-the-art baseline models. Our code is available at: https://github.com/downw/DT-GAT • We propose DT-GAT to integrate item- and session-level temporal signals. • Dual temporal GATs capture dependencies via temporal intra- and inter-session graphs. • Contrastive learning aligns dual channels to enhance session representations. • Experiments on three datasets validate the effectiveness of DT-GAT.

Knowledge-based Visual Question Answering (KB-VQA) extends general Visual Question Answering by requiring external knowledge beyond the provided visual and textual inputs, facilitating more complex real-world applications. KB-VQA introduces unique challenges, including the alignment of heterogeneous information from diverse modalities and sources, the retrieval of relevant knowledge from large-scale and noisy repositories, and the execution of complex reasoning to infer answers from the combined context. With the advancement of large language models, KB-VQA systems have undergone a notable transformation, where LLMs serve as powerful knowledge repositories, retrieval-augmented generators and strong reasoners. Despite substantial progress, there is a lack of a recent, systematic survey that organizes and reviews the evolving landscape of existing KB-VQA methods. This survey aims to fill this gap by establishing a structured taxonomy of KB-VQA approaches and decomposing mainstream systems into three fundamental stages: knowledge representation, knowledge retrieval, and knowledge reasoning. Through an examination of existing techniques employed at each stage, this survey identifies persistent challenges and outlines promising future research directions, providing a foundation for advancing KB-VQA models and their applications.

Implicit sentiment analysis (ISA) presents significant challenges due to the absence of salient cue words. Previous methods have struggled with insufficient data and limited reasoning capabilities to infer underlying opinions. Integrating multi-task learning (MTL) with large language models (LLMs) offers the potential to enable models of varying sizes to reliably perceive and recognize genuine opinions in ISA. However, existing MTL approaches are constrained by two sources of uncertainty:data-level uncertainty, arising from hallucination problems in LLM-generated contextual information, andtask-level uncertainty, stemming from the varying capacities of models to process contextual information. To handle these uncertainties, we proposeMT-ISA, a novel MTL framework that enhances ISA by leveraging the generation and reasoning capabilities of LLMs through automatic weight learning (AWL). Specifically,MT-ISAconstructs auxiliary tasks using generative LLMs to supplement sentiment elements and incorporates automatic MTL to fully exploit auxiliary data. We introduce data-level and task-level AWL, which dynamically identify relationships and prioritize more reliable data and critical tasks, enabling models of varying sizes to adaptively learn fine-grained weights based on their reasoning capabilities. Three strategies are investigated for data-level AWL, which are integrated with homoscedastic uncertainty for task-level AWL. Extensive experiments reveal that models of varying sizes achieve an optimal balance between primary prediction and auxiliary tasks inMT-ISA. This underscores the effectiveness and adaptability of our approach.

The challenges associated with large-scale user-item interaction graphs have attracted increasing attention in graph-based recommendation systems, primarily due to computational inefficiencies and inadequate information propagation. Existing methods provide partial solutions but suffer from notable limitations: model-centric approaches, such as sampling and aggregation, often struggle with generalization, while data-centric techniques, including graph sparsification and coarsening, lead to information loss and ineffective handling of bipartite graph structures. Recent advances in graph condensation offer a promising direction by reducing graph size while preserving essential information, presenting a novel approach to mitigating these challenges. Inspired by the principles of democracy, we proposeDemoRec, a framework that leverages graph condensation to generate user and item representatives for recommendation tasks. By constructing a compact interaction graph and clustering nodes with shared characteristics from the original graph, DemoRec significantly reduces graph size and computational complexity. Furthermore, it mitigates the over-reliance on high-order information, a critical challenge in large-scale bipartite graphs. Extensive experiments conducted on four public datasets demonstrate the effectiveness of DemoRec, showcasing substantial improvements in recommendation performance, computational efficiency, and robustness compared to SOTA methods.

Cross-Domain Recommendation (CDR) aims to alleviate data sparsity in the target domain by incorporating knowledge from external domains. Existing approaches typically rely on overlapping users between the source and target domains as a bridge for knowledge transfer. However, in practice, user information across domains is often unavailable due to privacy protection, platform isolation, and data sharing restrictions, rendering most methods ineffective. In this article, we propose the D2TCDR, a two-stage generative CDR framework to address this critical limitation. By modeling the domain-level distribution that captures user preferences shared across domains, we extract transferable knowledge and guide its transfer through a generative process, reducing reliance on overlapping users and alleviating data sparsity in the target domain. D2TCDR first proposes a domain disentanglement module to extract the domain-invariant representations, capturing shared preferences across domains by eliminating domain-specific interference. Subsequently, a guided diffusion model is designed to model the domain-level distribution of these domain-invariant representations. By injecting target-domain signals into the guided diffusion model, we further steer the learned distribution toward the target domain, achieving knowledge transfer without relying on overlapping users. Extensive experiments on multiple cross-domain datasets show the superior performance of D2TCDR, validating its recommendation capabilities in complex transfer scenarios. Code is available at: https://github.com/Red-Week/D2TCDR .

Recently, Large Language Models with Retrieval-Augmented Generation (RAG) have recently emerged as a powerful paradigm for sequential recommendation. However, existing methods typically retrieve items for each user without any principled mechanism for guaranteeing the reliability of generated recommendations, limiting their trustworthiness. To address this, we introduce SarRec : Statistically-guaranteed Augmented Retrieval for Recommendations, a framework that uses a simple retrieval step to provide relevant context and delivers calibrated, uncertainty-aware predictions with formal statistical guarantees. Specifically, SarRec first constructs the user's context set, utilizing a lightweight differentiable retrieval mechanism for identifying relevant context, and then calibrates the LLM's outputs by adapting the conformal prediction mechanism. We further provide a theoretical analysis that establishes an upper bound on the expected risk of recommendation performance metrics. Extensive experiments on multiple datasets from different domains validate the effectiveness of our framework.

Given that substantial amounts of domain-specific knowledge are stored in structured formats, such as web data organized through HTML, Large Language Models (LLMs) are expected to fully comprehend this structured information to broaden their applications in various real-world downstream tasks. Current approaches for applying LLMs to structured data fall into two main categories: serialization-based and operation-based methods. Both approaches, whether relying on serialization or using SQL-like operations as an intermediary, encounter difficulties in fully capturing structural relationships and effectively handling sparse data. To address these unique characteristics of structured data, we propose HyperG, a hypergraph-based generation framework aimed at enhancing LLMs' ability to process structured knowledge. Specifically, HyperG first augment sparse data with contextual information, leveraging the generative power of LLMs, and incorporate a prompt-attentive hypergraph learning (PHL) network to encode both the augmented information and the intricate structural relationships within the data. To validate the effectiveness and generalization of HyperG, we conduct extensive experiments across two different downstream tasks requiring structured knowledge. Our code is publicly available at: https://github.com/s1ruihuang/HyperG.

Knowledge graphs (KGs) consist of well-organized external information and have been proven to enhance recommendation quality effectively. Most KG-aware recommender systems are developed using real number space embeddings. In recent years, learning representations in the hypercomplex space has gained success and attention. Compared to single-component real-valued vectors, multi-component hypercomplex embeddings offer greater expressiveness, facilitating more meaningful modeling of users, items, entities, and their relations in the user-item interaction graph and KG. In this paper, we explore the integration of hypercomplex algebras in KG-aware recommendation and propose a Hypercomplex Knowledge Graph-aware Recommender (HKGR) method. Our HKGR models the interaction graph and KG in the hypercomplex space by utilizing specially designed hypercomplex graph neural networks. In particular, HKGR employs a hypercomplex attention-based aggregator to capture the structure and semantics of the KG. In the recommendation prediction phase, we design a hypercomplex interaction network that can approximate the high-order component interactions between users and items. Furthermore, we introduce a hypercomplex contrastive learning operator to strengthen cooperative signals between the interaction graph and KG modelings. Experiment results on the four real-world datasets show that our HKGR outperforms the state-of-the-art recommender baselines.

Abstract Large Language Models (LLMs) are widely applied in time series analysis. Yet, their utility in few-shot classification—a scenario with limited training data—remains unexplored. We aim to leverage the pre-trained knowledge in LLMs to overcome the data scarcity problem within multivariate time series. To this end, we propose LLMFew, an LLM-enhanced framework, to investigate the feasibility and capacity of LLMs for few-shot multivariate time series classification (MTSC). We first introduce a Patch-wise Temporal Convolution Encoder (PTCEnc) to align time series data with the textual embedding input of LLMs. Then, we fine-tune the pre-trained LLM decoder with Low-rank Adaptations (LoRA) to enable effective representation learning from time series data. Experimental results show our model consistently outperforms state-of-the-art baselines by a large margin, achieving 125.2% and 50.2% improvement in classification accuracy on Handwriting and EthanolConcentration datasets, respectively. Our results also show LLM-based methods achieve comparable performance to traditional models across various datasets in few-shot MTSC, paving the way for applying LLMs in practical scenarios where labeled data are limited. Our code is available at https://github.com/junekchen/llm-fewshot-mtsc .

E-commerce content generation necessitates creating engaging and customer-centric material to endorse products and enhance user satisfaction. Existing methods depend on task-specific feature design, which requires a fine-tailored model for each task with complex data collection and pre-processing, and their generation capabilities are limited. Meanwhile, large language models have demonstrated their capabilities in diverse natural language processing tasks, solving multiple tasks in a unified process. To address the concerns in e-commerce content generation, we leverage the impressive generation performance of large language models and propose a framework to educate them as proficient promoters in various e-commerce-related tasks. Our framework involves two modules:self-educatingproliferates task instructions and data by instructing the unaligned model, andmulti-aspect instruction alignmenteducates the language model by embedding all e-commerce tasks in a unified framework. The proposed model, Promoter, can perform a batch of prediction and generation tasks, working as a smart and creative promoter that only requires a quick view of the customer profile. Extensive experiments from automatic and human perspectives indicate that Promoter achieves state-of-the-art performances in various generation tasks, bringing the productivity of large language models to e-commerce in an integrated pipeline.

Fairness in recommendation has drawn much attention since it significantly affects how users access information and how information is exposed to users. However, most fairness-aware methods are designed offline with the entire stationary interaction data to handle the global unfairness issue and evaluate their performance in a one-time paradigm. In real-world scenarios, users tend to interact with items continuously over time, leading to a dynamic recommendation environment where unfairness is evolving online. Moreover, previous methods that focus on mitigating the unfairness can hardly bring significant improvements to the recommendation task. Hence, in this paper, we propose aModel-agnosticDual-sideOnlineFairness Learning method (MDOFair) for the dynamic recommendation. First, we carefully design dynamic dual-side fairness learning to trace the rapid evolution of unfairness from both the user and item sides. Second, we leverage the fairness and recommendation tasks in one utilized framework to pursue the double-win success. Last, we present an efficient model-agnostic post-ranking method for the dynamic recommendation scenario to mitigate the dynamic unfairness while improving the recommendation performance significantly. Extensive experiments demonstrate the superiority and effectiveness of our proposed MDOFair by incorporating it into existing dynamic models as a post-ranking stage.

Sequential recommender systems (SRSs) are designed to suggest relevant items to users by analyzing their interaction sequences. However, SRSs often suffer from exposure bias in these sequences due to imbalanced item exposure and varied user activity levels, creating a self-reinforcing loop favoring popular items regardless of their true relevance. Most SRSs only focus on item dependencies to address exposure bias, while overlooking user-side exposure bias and the rich semantics behind interactions. These oversights result in a limited understanding of less active users' preferences and inaccurate preference capture for less exposed items, exacerbating exposure biases. Towards this end, we propose a novel methodLLM-enhancedDualPropensity ScoreEstimation (LDPE), which synergistically integrates Large Language Models (LLMs) and causal inference. First, LDPE leverages LLMs' superior ability in capturing rich semantics from textual data and then integrates collaborative information to generate debiased semantic-rich LLM-based user/item embeddings. With these debiased item/user embeddings, LDPE estimates time-aware debiased propensity scores from both the item and user sides. These dual propensity scores can fully mitigate exposure bias by considering item popularity, user activity levels, and temporal dynamics. Lastly, LDPE employs the transformer as the backbone of our method, incorporating estimated dual propensity scores for accurately predicting users' true preferences. Extensive experiments show that our LDPE outperforms state-of-the-art baselines in terms of recommendation performance.

Conversational recommender systems (CRSs) provide personalised recommendations by strategically querying attributes matching users’ preferences. However, this process suffers from confounding effects of time and user attributes, as users’ preferences naturally evolve over time and differ among similar users due to their unique attributes. These confounding effects distort user behaviors’ causal drivers, challenging CRSs in learning users’ true preferences and generalizable patterns. Recently, causal inference provides principled tools to clarify cause-effect relations in data, offering a promising way to address such confounding effects. In this context, we introduceCausalConversationalRecommender (CCR), which applies causal inference to model the causality between user behaviors and time/user attribute, enabling deeper understanding of user behaviors’ causal drivers. First, CCR employs stratification and matching to ensure attribute asked per round is independent from time and user attributes, mitigating their confounding effects. Following that, we apply the Average Treatment Effect (ATE) to quantify the unbiased causal impact of each unasked attribute on user preferences, identifying the attribute with the highest ATE per round as the causal-based attribute, i.e., causal driver of user behaviour. Finally, CCR iteratively refines user preferences through feedback on causal-based attributes. Extensive experiments verified CCR's robustness and personalization.

With the explosive growth of online information, recommendation systems have become essential tools for alleviating information overload. In recent years, researchers have increasingly focused on centralized recommendation systems, capitalizing on the powerful computing capabilities of cloud servers and the rich historical data they store. However, the rapid development of edge computing and mobile devices in recent years has provided new alternatives for building recommendation systems. These alternatives offer advantages such as privacy protection and low-latency recommendations. To leverage the advantages of different computing nodes, including cloud servers, edge servers, and terminal devices, researchers have proposed recommendation systems that involve the collaboration of these nodes, known as distributed recommendation systems. This survey provides a systematic review of distributed recommendation systems. Specifically, we design a taxonomy for these systems from four perspectives and comprehensively summarize each study by category. In particular, we conduct a detailed analysis of the collaboration mechanisms of distributed recommendation systems. Finally, we discuss potential future research directions in this field.

Multi-Behavior Recommendation (MBR) aims to model personalized user preferences by integrating diverse interaction behaviors (e.g., page view, favorite, add to cart, purchase). However, latent confounders such as contextual influences and social relationships can obscure the true causal effects in real-world scenarios, thereby confounding the model’s prediction. Although existing MBR research extensively explores behavioral dependencies and heterogeneity, it frequently overlooks the impact of latent confounders, thereby limiting its ability to capture users’ genuine preferences. To address the limitations of existing methods, we identify two key challenges in MBR: (1) how to infer latent confounders, and (2) how to mitigate their influence across multi-behavior interactions. To this end, we propose Causal Variational Inference for Deconfounded (CVID) MBR. CVID employs a variational graph autoencoder to model latent uncertainty in multi-behavior interactions and introduces a confounder inference module to generate behavior-specific latent confounders via variational inference. In the conditional diffusion module, noise is progressively injected during the forward process to simulate the dynamic evolution of user preferences, while the reverse process leverages the inferred latent confounders to guide denoising through back-door adjustment, thereby recovering the true causal effects between multi-behavior interactions and the model’s prediction. Extensive experiments on public multi-behavior datasets demonstrate that CVID consistently outperforms state-of-the-art baselines in mitigating confounding effects and improving recommendation accuracy, validating its effectiveness and superiority.

Predicting user satisfaction over time is crucial in news recommendations, as users’ preferences are significantly influenced by various time-variant factors. Traditional correlation-based recommenders often suffer from redundant relationships, which can undermine their effectiveness over time. This work takes a time-aware causal approach to news recommendations, treating exposed news at a predicted time as the treatment variable and the resulting user satisfaction as the outcome variable. Capturing the evolving causal effects of exposed news items on user satisfaction poses significant challenges, particularly stemming from the need to model complex dependencies among time-variant covariates, such as news popularity and recency, as well as to effectively leverage the inherent user preferences embedded in time-invariant covariates. To these ends, we propose the CA u S al T ime-aware Rec ommender, named CAST-Rec , which accounts for the causal influences of both time-variant and time-invariant covariates. Specifically, we model the intricate causal dependencies among time-variant covariates through a series of transformer-based causal blocks. For time-invariant covariates, we utilize the semantic understanding and generative capabilities of Large Language Models (LLMs) to infer inherent user preferences while mitigating potential confounding effects. Extensive experiments demonstrate the superior performance of CAST-Rec compared to various news recommendation models and across multiple LLM implementations.

Multimodal recommendation, which utilizes rich multimodal information to learn user preferences, has attracted significant attention. Most works focus on designing powerful encoders for extracting multimodal features, and simply aggregate the learned features together to make prediction. Consequently, they have a limited capacity to learn the inter-modality knowledge including the modality-shared and modality-unique knowledge. In fact, learning the modality-shared knowledge enables us to align cross-modality data for fusing heterogeneous modality features. Learning the modality-unique knowledge is equally important when recommendation tasks only involve a small amount of shared features and the necessary information is contained within specific modality. In this article, we propose Contrastive Modality-Disentangled Learning (CMDL) to overcome this critical limitation. CMDL exactly captures the inter-modality knowledge by achieving modality disentanglement. Specifically, CMDL first disentangles the initial representation into the modality-invariant and modality-specific representations. Afterwards, CMDL introduces a novel manner of contrastive learning to approximate the MI upper bounds for achieving disentanglement regularization. Building upon the proposed regularization, CMDL encourages the modality-invariant and modality-specific representations to capture the modality-shared and modality-unique knowledge respectively and to be statistically independent to each other. Empirically, extensive experiments are conducted on benchmark datasets, demonstrating the superior performance of CMDL compared with strong multimodal recommenders.

Dynamic recommendation systems, where users interact with items continuously over time, have been widely deployed in real-world online streaming applications. The burst of interaction stream causes a rapid evolution of both users and items. To update representations dynamically, existing studies have investigated event-level and history-level dynamics by modeling the newly arrived interactions and aggregating historical interactions, respectively. However, most of them directly learn the representation evolution as new interactions occur, without exploring the collaboration between the newly arrived and historical interactions, thus failing to scrutinize whether those new interactions would benefit the evolution learning process when generating dynamic representations. Moreover, most of them model the two levels of dynamics independently, explicitly ignoring the inherent co-evolving correlation between them. In this work, we propose the Temporal Collaboration-Aware Graph Co-Evolution Learning (TCGC) for the dynamic recommendation scenario. First, we explore the effectiveness of collaborative information and devise the collaboration-aware indicator to guide the evolution learning process. Second, we design a temporal co-evolving graph network, enabling our framework to capture the correlation between event and history dynamics. Third, we leverage the evolution task and recommendation task together for joint training. Extensive experiments on four public datasets demonstrate the superiority and effectiveness of our proposed TCGC.

In social networks, echo chambers form when users primarily encounter information that reinforces their existing views with limited exposure to different perspectives. This self-reinforcing isolation worsens societal issues such as division and declining public discourse. Traditional approaches attempt to mitigate echo chambers by analyzing observable interaction patterns to identify their formative mechanisms. However, they overlook unobserved implicit factors, called hidden confounders in causal inference, that significantly influence content exposure and user behaviors despite not being directly captured in the data. To address this, we propose Causal Echo Diffusion Attenuator (CEDA) , a novel framework that integrates causal learning with sequential recommendations to detect and adjust for hidden confounders in social networks. Generally, CEDA comprises four key components: (1) User Dual Modelling builds comprehensive user embeddings by combining users’ attributes and structural information to fully capture behavior patterns. (2) Causal Transformer then estimates residual embeddings that account for hidden confounders, incorporating them into the Transformer as causal adjustments for unbiased user embeddings. (3) Social Diffusion Predictor uses unbiased user embeddings to jointly optimize diffusion prediction accuracy and information diversity. (4) Targeted Interventions strategically reshapes information flows to disrupt echo chambers based on the generated prediction and diversity insights. Extensive experiments demonstrate CEDA’s superior performance in both predicting information diffusion patterns and mitigating echo chambers.

Recently, there has been a surge in recommendations based on heterogeneous information networks (HINs), attributed to their ability to integrate complex and rich semantics. Despite this advancement, most HIN-based recommenders overlook two critical aspects. First, they often fail to consider HIN's heterophily nature, hindering the capture of non-local structures in HINs. Second, most methods lack the capability for logical reasoning. In this paper, we propose a novel structure- and logic-aware heterogeneous graph learning framework for recommender systems (SLHRec). Our SLHRec contains a structure-aware module and a logic-aware module. The former uses network geometry to construct non-local neighborhoods for nodes in HINs, and then introduces a graph neural network to integrate constructed neighbors for modeling the heterophily of HINs. The logic-aware module uses the Markov logic network (MLN) to infuse logic rules into heterogeneous graph learning, thereby boosting logic reasoning in recommendations. Furthermore, we utilize contrastive learning to model cooperative signals between modules, enabling them to complement each other. In the prediction stage, both modules contribute to generating recommendations. Compared with several strong recommender baselines, our SLHRec achieves superior performance on four real-world datasets.

Recent studies successfully learned static graph embeddings that are structurally fair by preventing the effectiveness disparity of high- and low-degree vertex groups in downstream graph mining tasks. However, achieving structure fairness in dynamic graph embedding remains an open problem. Neglecting degree changes in dynamic graphs will significantly impair embedding effectiveness without notably improving structure fairness. This is because the embedding performance of high-degree and low-to-high-degree vertices will significantly drop close to the generally poorer embedding performance of most slightly changed vertices in the long-tail part of the power-law distribution. We first identify biased structural evolutions in a dynamic graph based on the evolving trend of vertex degree and then propose FairDGE, the first structurally Fair Dynamic Graph Embedding algorithm. FairDGE learns biased structural evolutions by jointly embedding the connection changes among vertices and the long-short-term evolutionary trend of vertex degrees. Furthermore, a novel dual debiasing approach is devised to encode fair embeddings contrastively, customizing debiasing strategies for different biased structural evolutions. This innovative debiasing strategy breaks the effectiveness bottleneck of embeddings without notable fairness loss. Extensive experiments demonstrate that FairDGE achieves simultaneous improvement in the effectiveness and fairness of embeddings.

Product titles on e-commerce marketplaces often suffer from verbosity and inaccuracy, hindering effective communication of essential product details to customers. Refining titles to be more concise and informative is crucial for better user experience and product promotion. Recent solutions to product title refinement follow the standard text extractive and generative methods. Some also leverage multimodal information, e.g. using product images to supplement original titles with visual knowledge. However, these generative methods often produce additional terms not endorsed by sellers. Thus, it remains challenging to incorporate visual information missing from original titles into refined titles without excessively introducing novel terms. Additionally, most existing methods require human-labeled datasets, which are laborious to construct. In response to the two challenges, we present a self-supervised multimodal framework (HLATR) for title refinement that comprises two key modules: (1) a perturbated sample generator that constructs training data by systematically mining homogeneous listing information and (2) a title refinement network that effectively harnesses visual information to refine the original titles. To explicitly balance the extraction from original titles and the generation of supplementary novel terms, we adapt the copy mechanism that is guided by a focused refinement loss. Extensive experiments demonstrate that our proposed framework consistently outperforms others in generating refined titles that contain essential multimodal semantics with minimal deviation from the original ones.

Language models such as Bidirectional Encoder Representations from Transformers (BERT) have been very effective in various Natural Language Processing (NLP) and text mining tasks including text classification. However, some tasks still pose challenges for these models, including text classification with limited labels. This can result in a cold-start problem. Although some approaches have attempted to address this problem through single-stage clustering as an intermediate training step coupled with a pre-trained language model, which generates pseudo-labels to improve classification, these methods are often error-prone due to the limitations of the clustering algorithms. To overcome this, we have developed a novel two-stage intermediate clustering with subsequent fine-tuning that models the pseudo-labels reliably, resulting in reduced prediction errors. The key novelty in our model, IDoFew, is that the two-stage clustering coupled with two different clustering algorithms helps exploit the advantages of the complementary algorithms that reduce the errors in generating reliable pseudo-labels for fine-tuning. Our approach has shown significant improvements compared to strong comparative models.

Social relations are leveraged to tackle the sparsity issue of user-item interaction data in recommendation under the assumption of social homophily. However, social recommendation paradigms predominantly focus on homophily based on user preferences. While social information can enhance recommendations, its alignment with user preferences is not guaranteed, thereby posing the risk of introducing informational redundancy. We empirically discover that social graphs in real recommendation data exhibit low preference-aware homophily, which limits the effect of social recommendation models. To comprehensively extract preference-aware homophily information latent in the social graph, we propose Social Heterophily-alleviating Rewiring (SHaRe), a data-centric framework for enhancing existing graph-based social recommendation models. We adopt Graph Rewiring technique to capture and add highly homophilic social relations, and cut low homophilic (or heterophilic) relations. To better refine the user representations from reliable social relations, we integrate a contrastive learning method into the training of SHaRe, aiming to calibrate the user representations for enhancing the result of Graph Rewiring. Experiments on real-world datasets show that the proposed framework not only exhibits enhanced performances across varying homophily ratios but also improves the performance of existing state-of-the-art (SOTA) social recommendation models.

Graph collaborative filtering (GCF) has emerged as a prominent method in recommendation systems, leveraging the power of graph learning to enhance traditional collaborative filtering (CF). One common approach in GCF involves employing Graph Convolutional Networks (GCN) to learn user and item embeddings and utilize these embeddings to optimize CF models. However, existing GCN-based methods often fall short of generating satisfactory embeddings, mainly due to their limitations in capturing node dependencies and variable dependencies within the graph. Consequently, the learned embeddings are fragile in uncovering the root causes of user preferences, leading to sub-optimal performance of GCF models. In this work, we propose integrating causal modeling with the learning process of GCN-based GCF models, leveraging causality-aware graph embeddings to capture complex dependencies in recommendations. Our methodology encompasses three key designs: 1) Causal Graph conceptualization, 2) Neural Causal Model parameterization, and 3) Variational inference for the Neural Causal Model. We define a Causal Graph to model genuine dependencies in GCF models and utilize this Causal Graph to parameterize a Neural Causal Model. The proposed framework, termed Neural Causal Graph Collaborative Filtering (NCGCF), uses variational inference to approximate neural networks under the Neural Causal Model. As a result, NCGCF is able to leverage the expressive causal effects from the Causal Graph to enhance graph representation learning. Extensive experimentation on four datasets demonstrates NCGCF's ability to deliver precise recommendations consistent with user preferences.

The Natural Language to Visualization (NL2Vis) task aims to transform natural-language descriptions into visual representations for a grounded table, enabling users to gain insights from vast amounts of data. Recently, many deep learning-based approaches have been developed for NL2Vis. Despite the considerable efforts made by these approaches, challenges persist in visualizing data sourced from unseen databases or spanning multiple tables. Taking inspiration from the remarkable generation capabilities of Large Language Models (LLMs), this paper conducts an empirical study to evaluate their potential in generating visualizations, and explore the effectiveness of in-context learning prompts for enhancing this task. In particular, we first explore the ways of transforming structured tabular data into sequential text prompts, as to feed them into LLMs and analyze which table content contributes most to the NL2Vis. Our findings suggest that transforming structured tabular data into programs is effective, and it is essential to consider the table schema when formulating prompts. Furthermore, we evaluate two types of LLMs: finetuned models (e.g., T5-Small) and inference-only models (e.g., GPT-3.5), against state-of-the-art methods, using the NL2Vis benchmarks (i.e., nvBench). The experimental results reveal that LLMs outperform baselines, with inference-only models consistently exhibiting performance improvements, at times even surpassing fine-tuned models when provided with certain few-shot demonstrations through in-context learning. Finally, we analyze when the LLMs fail in NL2Vis, and propose to iteratively update the results using strategies such as chain-of-thought, role-playing, and code-interpreter. The experimental results confirm the efficacy of iterative updates and hold great potential for future study.

Recommender system helps address information overload problem and satisfy consumers’ personalized requirement in many applications such as e-commerce, social networks, and in-game store. However, existing approaches mainly focus on improving the accuracy of recommendation tasks but usually ignore how to improve the interpretability of recommendation, which is still a challenging and crucial task, especially for some complicated scenarios such as large-scale online games. A few previous attempts on explainable recommendation mostly depend on a large amount of a priori knowledge or user-provided review corpus, which is labor consuming as well as often suffers from data deficiency. To relieve this issue, we propose a Multi-source Heterogeneous Graph Attention Network for Explainable Recommendation (MHANER) for the case without enough a priori knowledge or corpus of user comments. Specifically, MHANER employs the attention mechanism to model players’ preference to in-game store items as the support for the explanation of recommendation. Then a graph neural network–based method is designed to model players’ multi-source heterogeneous information, including the players’ historical behavior data, historical purchase data, and attributes of the player-controlled character, which is leveraged to recommend possible items for players to buy. Finally, the multi-level subgraph pattern mining is adopted to combine the characteristics of a recommendation list to generate corresponding explanations of items. Extensive experiments on three real-world datasets, two collected from JD and one from NetEase game, demonstrate that the proposed model MHANER outperforms state-of-the-art baselines. Moreover, the generated explanations are verified by human encoding comprised of hard-core game players and endorsed by experts from game developers.

With the wide application of recommender systems, the potential impacts of recommender systems on customers, item providers and other parties have attracted increasing attention. Fairness, which is the quality of treating people equally, is also becoming important in recommender system evaluation and algorithm design. Therefore, in the past years, there has been a growing interest in fairness measurement and assurance in recommender systems. Although there are several reviews on related topics, such as fairness in machine learning and debias in recommender systems, they do not present a systematic view on fairness in recommender systems, which is context aware and has a multi-sided meaning. Therefore, in this review, the concept of fairness is discussed in detail in the various contexts of recommender systems. Specifically, a comprehensive framework to classify fairness metrics is proposed from four dimensions, i.e.,Fairness for Whom,Demographic Unit,Time Frame, andQuantification Method. Then the strategies for eliminating unfairness in recommendations, fairness in different recommendation tasks and datasets are reviewed and summarized. Finally, the challenges and future work are discussed.

Compared with only pursuing recommendation accuracy, the explainability of a recommendation model has drawn more attention in recent years. Many graph-based recommendations resort to informative paths with the attention mechanism for the explanation. Unfortunately, these attention weights are intentionally designed for model accuracy but not explainability. Recently, some researchers have started to question attention-based explainability because the attention weights are unstable for different reproductions, and they may not always align with human intuition. Inspired by the counterfactual reasoning from causality learning theory, we propose a novel explainable framework targeting path-based recommendations, wherein the explainable weights of paths are learned to replace attention weights. Specifically, we design two counterfactual reasoning algorithms from both path representation and path topological structure perspectives. Moreover, unlike traditional case studies, we also propose a package of explainability evaluation solutions with both qualitative and quantitative methods. We conduct extensive experiments on four real-world datasets, the results of which further demonstrate the effectiveness and reliability of our method.

Counterfactual explanations interpret the recommendation mechanism by exploring how minimal alterations on items or users affect recommendation decisions. Existing counterfactual explainable approaches face huge search space, and their explanations are either action-based (e.g., user click) or aspect-based (i.e., item description). We believe item attribute-based explanations are more intuitive and persuadable for users since they explain by fine-grained demographic features, e.g., brand. Moreover, counterfactual explanations could enhance recommendations by filtering out negative items. In this work, we propose a novelCounterfactual Explainable Recommendation (CERec)to generate item attribute-based counterfactual explanations meanwhile to boost recommendation performance. OurCERecoptimizes an explanation policy upon uniformly searching candidate counterfactuals within a reinforcement learning environment. We reduce the huge search space with an adaptive path sampler by using rich context information of a given knowledge graph. We also deploy the explanation policy to a recommendation model to enhance the recommendation. Extensive explainability and recommendation evaluations demonstrateCERec's ability to provide explanations consistent with user preferences and maintain improved recommendations. We release our code and processed datasets athttps://github.com/Chrystalii/CERec.

Graph Contrastive Learning (GCL) stands as a potent framework for unsupervised graph representation learning that has gained traction across numerous graph learning applications. The effectiveness of GCL relies on generating high-quality contrasting samples, enhancing the model’s ability to discern graph semantics. However, the prevailing GCL methods face two key challenges: 1) introducing noise during graph augmentations and 2) requiring additional storage for generated samples, which degrade the model performance. In this paper, we propose novel approaches, GKCL (i.e., Graph Knowledge Contrastive Learning) and DGKCL (i.e., Distilled Graph Knowledge Contrastive Learning), that leverage multi-level graph knowledge to create noise-free contrasting pairs. This framework not only addresses the noise-related challenges but also circumvents excessive storage demands. Furthermore, our method incorporates a knowledge distillation component to optimize the trained embedding tables, reducing the model’s scale while ensuring superior performance, particularly for the scenarios with smaller embedding sizes. Comprehensive experimental evaluations on three public benchmark datasets underscore the merits of our proposed method and elucidate its properties, which primarily reflect the performance of the proposed method equipped with different embedding sizes and how the distillation weight affects the overall performance.

Conversational Recommender Systems (CRSs) fundamentally differ from traditional recommender systems by interacting with users in a conversational session to accurately predict their current preferences and provide personalized recommendations. Although current CRSs have achieved favorable recommendation performance, the explainability is still in its infancy stage. Most of the CRSs tend to provide coarse explanations and fail to explore the impact of minimal alterations on the recommendation decisions on items. In this paper, we are the first to incorporate the counterfactual techniques into CRS and propose a Counterfactual Explainable Conversational Recommender (CECR) to enhance the recommendation model from a counterfactual perspective. Counterfactual explanations can offer fine-grained reasons to explain users' real-time intentions, meanwhile generating counterfactual samples for augmenting the training dataset to enhance recommendation performance. Specifically, CECR adaptively learns users' preferences based on the conversation context and effectively responds to users' real-time feedback during multiple rounds of conversation. Furthermore, CECR actively generates counterfactual samples to augment the training set and thus leading to a constant improvement in recommendation performance. Empirical experiments carried out on three benchmark datasets show that our CECR outperforms state-of-the-art CRSs in terms of recommendation performance and explainability

Sequential prediction has great value for resource allocation due to its capability in analyzing intents for next prediction. A fundamental challenge arises from real-world interaction dynamics where similar sequences involving multiple intents may exhibit different next items. More importantly, the character of volume candidate items in sequential prediction may amplify such dynamics, making deep networks hard to capture comprehensive intents. This article presents a sequential prediction framework with Decoupled Progressive Distillation (DePoD), drawing on the progressive nature of human cognition. We redefine target and non-target item distillation according to their different effects in the decoupled formulation. This can be achieved through two aspects: (1) Regarding how to learn, our target item distillation with progressive difficulty increases the contribution of low-confidence samples in the later training phase while keeping high-confidence samples in the earlier phase. And, the non-target item distillation starts from a small subset of non-target items from which size increases according to the item frequency. (2) Regarding whom to learn from, a difference evaluator is utilized to progressively select an expert that provides informative knowledge among items from the cohort of peers. Extensive experiments on four public datasets show DePoD outperforms state-of-the-art methods in terms of accuracy-based metrics.

Fairness-aware recommendation alleviates discrimination issues to build trustworthy recommendation systems. Explaining the causes of unfair recommendations is critical, as it promotes fairness diagnostics, and thus secures users’ trust in recommendation models. Existing fairness explanation methods suffer high computation burdens due to the large-scale search space and the greedy nature of the explanation search process. Besides, they perform feature-level optimizations with continuous values, which are not applicable to discrete attributes such as gender and age. In this work, we adopt counterfactual explanations from causal inference and propose to generate attribute-level counterfactual explanations, adapting to discrete attributes in recommendation models. We use real-world attributes from Heterogeneous Information Networks (HINs) to empower counterfactual reasoning on discrete attributes. We propose a Counterfactual Explanation for Fairness (CFairER) that generates attribute-level counterfactual explanations from HINs for item exposure fairness. Our CFairER conducts off-policy reinforcement learning to seek high-quality counterfactual explanations, with attentive action pruning reducing the search space of candidate counterfactuals. The counterfactual explanations help to provide rational and proximate explanations for model fairness, while the attentive action pruning narrows the search space of attributes. Extensive experiments demonstrate our proposed model can generate faithful explanations while maintaining favorable recommendation performance.

Traditional music recommender systems are mainly based on users’ interactions, which limit their performance. Particularly, various kinds of content information, such as metadata and description can be used to improve music recommendation. However, it remains to be addressed how to fully incorporate the rich auxiliary/side information and effectively deal with heterogeneity in it. In this paper, we propose a M ulti-view E nhanced G raph A ttention N etwork (named MEGAN ) for session-based music recommendation. MEGAN can learn informative representations (embeddings) of music pieces and users from heterogeneous information based on graph neural network and attention mechanism. Specifically, the proposed approach MEGAN firstly models users’ listening behaviors and the textual content of music pieces with a Heterogeneous Music Graph (HMG). Then, a devised Graph Attention Network is used to learn the low-dimensional embedding of music pieces and users and by integrating various kinds of information, which is enhanced by multi-view from HMG in an adaptive and unified way. Finally, users’ hybrid preferences are learned from users’ listening behaviors and music pieces that satisfy users real-time requirements are recommended. Comprehensive experiments are conducted on two real-world datasets, and the results show that MEGAN achieves better performance than baselines, including several state-of-the-art recommendation methods.

When only users’ preferences and interests are considered by a recommendation algorithm, it will lead to the severe long-tail problem over items. Therefore, the unfair exposure phenomenon of recommended items caused by this problem has attracted widespread attention in recent years. For the first time, we reveal the fact that there is a more serious unfair exposure problem in session-based recommender systems (SRSs), which learn the short-term and dynamic preferences of users from anonymous sessions. Considering the fact that in SRSs, recommendations are provided multiple times and item exposures are accumulated over interactions in a session, we define new metrics both for the fairness of item exposure and recommendation quality among sessions. Moreover, we design a dynamic F airness- A ssurance ST rategy for s E ssion-based R ecommender systems ( FASTER ). FASTER is a post-processing strategy that tries to keep a balance between item exposure fairness and recommendation quality. It can also maintain the fairness of recommendation quality among sessions. The effectiveness of FASTER is verified on three real-world datasets and five original algorithms. The experiment results show that FASTER can generally reduce the unfair exposure of different session-based recommendation algorithms while still ensuring a high level of recommendation quality.

Fairness-aware recommendation eliminates discrimination issues to build trustworthy recommendation systems. Existing fairness-aware approaches ignore accounting for rich user and item attributes and thus cannot capture the impact of attributes on affecting recommendation fairness. These real-world attributes severely cause unfair recommendations by favoring items with popular attributes, leading to item exposure unfairness in recommendations. Moreover, existing approaches mostly mitigate unfairness for static recommendation models, e.g., collaborative filtering. Static models can not handle dynamic user interactions with the system that reflect users’ preferences shift through time. Thus, static models are limited in their ability to adapt to user behavior shifts to gain long-run user satisfaction. As user and item attributes are largely involved in modern recommenders and user interactions are naturally dynamic, it is essential to develop a novel method that eliminates unfairness caused by attributes meanwhile embrace the dynamic modeling of user behavior shifts. In this article, we propose Constrained Off-policy Learning over Heterogeneous Information for Fairness-aware Recommendation (Fair-HINpolicy) , which uses recent advances in context-aware off-policy learning to produce fairness-aware recommendations with rich attributes from a Heterogeneous Information Network. In particular, we formulate the off-policy learning as a Constrained Markov Decision Process (CMDP) by dynamically constraining the fairness of item exposure at each iteration. We also design an attentive action sampling to reduce the search space for off-policy learning. Our solution adaptively receives HIN-augmented corrections for counterfactual risk minimization, and ultimately yields an effective policy that maximizes long-term user satisfaction. We extensively evaluate our method through simulations on large-scale real-world datasets, obtaining favorable results compared with state-of-the-art methods.

On-device recommender systems recently have garnered increasing attention due to their advantages of providing prompt response and securing privacy. To stay current with evolving user interests, cloud-based recommender systems are periodically updated with new interaction data. However, on-device models struggle to retrain themselves because of limited onboard computing resources. As a solution, we consider the scenario where the model retraining occurs on the server side and then the updated parameters are transferred to edge devices via network communication. While this eliminates the need for local retraining, it incurs a regular transfer of parameters that significantly taxes network bandwidth. To mitigate this issue, we develop an efficient approach based on compositional codes to compress the model update. This approach ensures the on-device model is updated flexibly with minimal additional parameters whilst utilizing previous knowledge. The extensive experiments conducted on multiple session-based recommendation models with distinctive architectures demonstrate that the on-device model can achieve comparable accuracy to the retrained server-side counterpart through transferring an update 60x smaller in size. The codes are available at https://github.com/xiaxin1998/ODUpdate.

Temporal knowledge graph (TKG) extrapolation aims to predict future unknown events (facts) based on historical information, and has attracted considerable attention due to its great practical significance. Accurate representations (embeddings) of entities and relations form the basis of TKG extrapolation. Recent work has been devoted to improving the rationality of entity representations. However, on the one hand, ignoring relation modeling results in incomplete relation representations; therefore, some approaches aggregate only immediately adjacent entities of relations, but this can lead to the "message islands" problem of relation modeling. On the other hand, ignoring the association constraints between relations and entities can make the embeddings of both relations and entities prone to overfitting. To address the abovementioned challenges, we propose an advanced method, namely, RETIA. For the former issue, we generate twin hyperrelation subgraphs for each historical subgraph and then aggregate both the adjacent entities and relations in the hyperrelation subgraphs through a graph convolutional network (GCN). About the latter concern, we propose a twin-interact module (TIM), which provides communication channels for relation aggregation and entity aggregation during the evolution of the historical sequence. Experiments conducted on five public datasets show that RETIA has made great improvements across several evaluation metrics. Our released code is available at https://github.com/CGCL-codes/RETIA.

Temporal knowledge graph (TKG) forecasting is widely used in various fields due to its ability to infer future events based on historical information. Modeling the internal structures and chronological dependencies of historical subgraph sequences has been proven effective. Nevertheless, on the one hand, the TKG forecasting process generally suffers from a lack of sufficient sample data due to historical resource limitations; thus, most works focus on continuously mining the patterns of historical sequences while ignoring the semantically-rich background information provided by external knowledge, especially when historical query-related information is scarce. On the other hand, when merely serializing the given subgraph sequence to mimic its temporal evolution process, only the chronological dependencies between the subgraphs can be considered, thus ignoring the evolution of time information. Hence, a method that integrates internal and external knowledge to enhance the representations of entities is urgently needed. To this end, we propose a novel TKG forecasting method, namely, the internal and external evolution-enhanced framework (IE-Evo). For the former issue, we design an external evolution encoder and use a pre-trained language model (PLM) to provide powerful external knowledge semantics for TKG forecasting. To address the latter concern, we propose an internal evolution encoder that explicitly embeds the time information while modeling the aggregation and evolution processes of the observed sequential structural information. IE-Evo has been evaluated on four public benchmark datasets, showcasing its significant improvements across multiple evaluation metrics.

Improving users’ long-term experience in recommender systems (RS) has become a growing concern for recommendation platforms. Reinforcement learning (RL) is an attractive approach because it can plan and optimize long-term returns sequentially. However, directly applying RL as an online learning method in the RS setting can significantly compromise users’ satisfaction and experience. As a result, learning the recommendation policy from logged feedback collected under different policies has emerged as a promising direction. Offline learning enables the agent to utilize off-policy learning techniques. Nevertheless, several challenges need to be addressed, such as distribution shift. In this paper, we propose a novel RL method, called Supervised Off-Policy Actor-Critic (SOAC), for learning the recommendation policy from the logged feedback without exploration. The proposed SOAC addresses challenges, including distribution shift and extrapolation errors, and focuses on improving the ranking of items in a recommendation list. The experimental results demonstrate that SOAC can achieve better recommendation performance than existing supervised RL methods.

Abstract Emotion recognition (ER) from speech signals is a robust approach since it cannot be imitated like facial expression or text based sentiment analysis. Valuable information underlying the emotions are significant for human-computer interactions enabling intelligent machines to interact with sensitivity in the real world. Previous ER studies through speech signal processing have focused exclusively on associations between different signal mode decomposition methods and hidden informative features. However, improper decomposition parameter selections lead to informative signal component losses due to mode duplicating and mixing. In contrast, the current study proposes VGG-optiVMD, an empowered variational mode decomposition algorithm, to distinguish meaningful speech features and automatically select the number of decomposed modes and optimum balancing parameter for the data fidelity constraint by assessing their effects on the VGG16 flattening output layer. Various feature vectors were employed to train the VGG16 network on different databases and assess VGG-optiVMD reproducibility and reliability. One, two, and three-dimensional feature vectors were constructed by concatenating Mel-frequency cepstral coefficients, Chromagram, Mel spectrograms, Tonnetz diagrams, and spectral centroids. Results confirmed a synergistic relationship between the fine-tuning of the signal sample rate and decomposition parameters with classification accuracy, achieving state-of-the-art 96.09% accuracy in predicting seven emotions on the Berlin EMO-DB database.

As an indispensable personalized service in Location-based Social Networks (LBSNs), the next Point-of-Interest (POI) recommendation aims to help people discover attractive and interesting places. Currently, most POI recommenders are based on the conventional centralized paradigm that heavily relies on the cloud to train the recommendation models with large volumes of collected users' sensitive check-in data. Although a few recent works have explored on-device frameworks for resilient and privacy-preserving POI recommendations, they invariably hold the assumption of model homogeneity for parameters/gradients aggregation and collaboration. However, users' mobile devices in the real world have various hardware configurations (e.g., compute resources), leading to heterogeneous on-device models with different architectures and sizes. In light of this, We propose a novel on-device POI recommendation framework, namely Model-Agnostic Collaborative learning for on-device POI recommendation (MAC), allowing users to customize their own model structures (e.g., dimension & number of hidden layers). To counteract the sparsity of on-device user data, we propose to pre-select neighbors for collaboration based on physical distances, category-level preferences, and social networks. To assimilate knowledge from the above-selected neighbors in an efficient and secure way, we adopt the knowledge distillation framework with mutual information maximization. Instead of sharing sensitive models/gradients, clients in MAC only share their soft decisions on a preloaded reference dataset. To filter out low-quality neighbors, we propose two sampling strategies, performance-triggered sampling and similarity-based sampling, to speed up the training process and obtain optimal recommenders. In addition, we design two novel approaches to generate more effective reference datasets while protecting users' privacy. Extensive experiments on two datasets have shown the superiority of MAC over advanced baselines.

Graph neural network (GNN) based algorithms have achieved superior performance in recommendation tasks due to their advanced capability of exploiting high-order connectivity between users and items. However, most existing GNN-based recommendation models ignore the dynamic evolution of nodes, where users will continuously interact with items over time, resulting in rapid changes in the environment (e.g., neighbor and structure). Moreover, the heuristic normalization of embeddings in dynamic recommendation is de-coupled with the model learning process, making the whole system suboptimal. In this paper, we propose a novel framework for generating satisfying recommendations in dynamic environments, called Dynamic Graph Evolution Learning (DGEL). First, we design three efficient real-time update learning methods for nodes from the perspectives of inherent interaction potential, time-decay neighbor augmentation, and symbiotic local structure learning. Second, we construct the re-scaling enhancement networks for dynamic embeddings to adaptively and automatically bridge the normalization process with model learning. Third, we leverage the interaction matching task and the future prediction task together for joint training to further improve performance. Extensive experiments on three real-world datasets demonstrate the effectiveness and improvements of our proposed DGEL. The code is available at https://github.com/henrictang/DGEL.

Graph Convolutional Networks (GCNs) are a popular type of machine learning models that use multiple layers of convolutional aggregation operations and non-linear activations to represent data. Recent studies apply GCNs to Collaborative Filtering (CF)-based recommender systems (RSs) by modeling user-item interactions as a bipartite graph and achieve superior performance. However, these models face difficulty in training with non-linear activations on large graphs. Besides, most GCN-based models could not model deeper layers due to the over-smoothing effect with the graph convolution operation. In this paper, we improve the GCN-based CF models from two aspects. First, we remove non-linearities to enhance recommendation performance, which is consistent with the theories in simple graph convolutional networks. Second, we obtain the initialization of the embedding for each node in the graph by computing the network embedding on the condensed graph, which alleviates the over smoothing problem in graph convolution aggregation operation with sparse interaction data. The proposed model is a linear model that is easy to train, scalable to large datasets, and shown to yield better efficiency and effectiveness on four real datasets.

The session-based recommendation aims to predict users' immediate next actions based on their short-term behaviors reflected by past and ongoing sessions. Graph neural networks (GNNs) recently dominated the related studies, yet their performance heavily relies on graph structures, which are often predefined, task-specific, and designed heuristically. Furthermore, existing graph-based methods either neglect implicit correlations among items or consider explicit and implicit relationships altogether in the same graphs. We propose to decouple explicit and implicit relationships among items. As such, we can capture the prior knowledge encapsulated in explicit dependencies and learned implicit correlations among items simultaneously in a flexible and more interpretable manner for effective recommendations. We design a dual graph neural network that leverages the feature representations extracted by two GNNs: a graph neural network with a single gate (SG-GNN) and an adaptive graph neural network (A-GNN). The former models explicit dependencies among items. The latter employs a self-learning strategy to capture implicit correlations among items. Our experiments on four real-world datasets show our model outperforms state-of-the-art methods by a large margin, achieving 18.46% and 70.72% improvement in [email protected], and 49.10% and 115.29% improvement in [email protected] on Diginetica and LastFM datasets.

Graph contrastive learning has emerged as a powerful unsupervised graph representation learning tool. The key to the success of graph contrastive learning is to acquire high-quality positive and negative samples as contrasting pairs to learn the underlying structural semantics of the input graph. Recent works usually sample negative samples from the same training batch with the positive samples or from an external irrelevant graph. However, a significant limitation lies in such strategies: the unavoidable problem of sampling false negative samples. In this paper, we propose a novel method to utilize Counterfactual mechanism to generate artificial hard negative samples for Graph Contrastive learning, namely CGC. We utilize a counterfactual mechanism to produce hard negative samples, ensuring that the generated samples are similar but have labels that differ from the positive sample. The proposed method achieves satisfying results on several datasets. It outperforms some traditional unsupervised graph learning methods and some SOTA graph contrastive learning methods. We also conducted some supplementary experiments to illustrate the proposed method, including the performances of CGC with different hard negative samples and evaluations for hard negative samples generated with different similarity measurements. The implementation code is available online to ease reproducibility1.

In recommendation systems, the existence of the missing-not-at-random (MNAR) problem results in the selection bias issue, degrading the recommendation performance ultimately. A common practice to address MNAR is to treat missing entries from the so-called “exposure” perspective, i.e., modeling how an item is exposed (provided) to a user. Most of the existing approaches use heuristic models or re-weighting strategy on observed ratings to mimic the missing-at-random setting. However, little research has been done to reveal how the ratings are missing from a causal perspective. To bridge the gap, we propose an unbiased and robust method called DENC ( De-Bias Network Confounding in Recommendation ), inspired by confounder analysis in causal inference. In general, DENC provides a causal analysis on MNAR from both the inherent factors (e.g., latent user or item factors) and auxiliary network’s perspective. Particularly, the proposed exposure model in DENC can control the social network confounder meanwhile preserve the observed exposure information. We also develop a deconfounding model through the balanced representation learning to retain the primary user and item features, which enables DENC generalize well on the rating prediction. Extensive experiments on three datasets validate that our proposed model outperforms the state-of-the-art baselines.

Recent advances in path-based explainable recommendation systems have attracted increasing attention thanks to the rich information from knowledge graphs. Most existing explainable recommendations only utilize static knowledge graphs and ignore the dynamic user-item evolutions, leading to less convincing and inaccurate explanations. Although some works boost the performance and explainability of recommendations through modeling the user's temporal sequential behavior, most of them either only focus on modeling the user's sequential interactions within a path or independently and separately of the recommendation mechanism. Moreover, some path-based explainable recommendations use random selection or traditional machine learning methods to decrease the volume of explainable paths, which cannot guarantee high quality of the explainable paths for the recommendation. To deal with the problem, recent path exploration use reinforcement learning to improve diversity and quality. However, unsupervised training leads to low-efficiency path exploration. Therefore, we propose a novelTemporalMeta-path GuidedExplainableRecommendation leveragingReinforcementLearning (TMER-RL), which utilizes supervised reinforcement learning to explore item-item paths between consecutive items with attention mechanisms to sequentially model dynamic user-item evolutions on a dynamic knowledge graph for the explainable recommendation. Extensive evaluations of TMER-RL on two real-world datasets show state-of-the-art performance compared to recent strong baselines.

Recently, deep learning techniques have yielded immense success on recommender systems. However, one weakness of most deep methods is that, users/items mutual semantic relationships, which are latent in the user-item interactions, are not distilled out explicitly. Moreover, most methods have been primarily focused on representation learning in euclidean geometry. Since recent studies have shown that the bipartite graph structure has the non-euclidean latent anatomy, euclidean embeddings may suffer from a certain degree of distortion. In this work, we presentHyperbolicNeuralCollaborativeRecommender (HNCR), a deep hyperbolic representation learning method that exploits mutual semantic relationships among users/items for collaborative filtering tasks. HNCR first introduces a neighbor construction strategy to build user and item semantic neighborhoods. Then HNCR develops a framework based on deep learning and hyperbolic geometry to integrate constructed neighborhoods into recommendation. To evaluate our method, we conduct experiments on the four datasets. Experimental results show the superiority of HNCR compared with its euclidean counterpart and state-of-the-art recommendation baselines. The results also indicate that hyperbolic representations can reflect meaningful data insights.

Modern sociology has profoundly uncovered many convincing social criteria for behavioral analysis. Unfortunately, many of them are too subjective to be measured and very challenging to be presented in online social networks (OSNs) for the large data volume and complicated environments to be explored. On the other hand, data mining techniques can better find data patterns but many of them leave behind unnatural understanding to humans. Although there are some works trying to integrate social observations for specific tasks, they are still hard to be applied to more general cases. In this paper, we propose a fundamental methodology to support the further fusion of data mining techniques and sociological behavioral criteria. Our highlights are three-fold: First, we propose an effective hypergraph awareness and a fast line graph construction framework. The hypergraph can more profoundly indicate the interactions between individuals and their environments because each edge in the hypergraph (a.k.a hyperedge) contains more than two nodes, which is perfect to describe social. A line graph treats each social environment as a super node with the underlying influence between different environments. In this way, we go beyond traditional pair-wise relations and explore richer patterns under various sociological criteria; Second, we propose a novel hypergraph-based neural network to learn social influence flowing from users to users, users to environments, environment to users, and environments to environments. The neural network can be learned via a task-free method, making our model very flexible to support various data mining tasks and sociological analysis; Third, we propose both qualitative and quantitive solutions to effectively evaluate the most common sociological criteria like social conformity, social equivalence, environmental evolving and social polarization. Our extensive experiments show that our framework can better support both data mining tasks for online user behaviors and sociological analysis.

Traditional recommendation models trained on observational interaction data have generated large impacts in a wide range of applications, it faces bias problems that cover users’ true intent and thus deteriorate the recommendation effectiveness. Existing methods track this problem as eliminating bias for the robust recommendation, e.g., by re-weighting training samples or learning disentangled representations. The disentangled representation methods as the state-of-the-art eliminate bias by revealing cause-effect of the bias generation. However, how to design the semantic-aware and unbiased representations for users’ true intents is largely unexplored. To bridge the gap, we are the first to propose an unbiased and semantic-aware disentanglement learning calledCaDSI(CausalDisentanglement forSemantics-AwareIntent Learning) from a causal perspective. Particularly, CaDSI explicitly models the causal relations underlying recommendation task, and thus produces semantic-aware representations via disentangling users’ true intents aware of specific item context. Moreover, the causal intervention mechanism is designed to eliminate confounding bias stemming from context information, which further aligns the semantic-aware representation with users’ true intent. Extensive experiments and case studies both validate the robustness and interpretability of our proposed model.

Recommendation and search systems have already become indispensable means for helping web users identify the most relevant information/services in the era of information overload. The applications of such systems are multi-faceted, including targeted advertising, intelligent medical assistant, and e-commerce, and are bringing immense convenience to people’s daily lives. However, despite rapid advances in recommendation and search, the increasing public awareness of the trustworthiness of relevant recommendation and search applications has introduced higher expectations on relevant research. Firstly, the unprecedentedly growing heterogeneity of use cases has been challenging the adaptivity of contemporary algorithms to various settings, e.g., dynamic user interests [Chen et al. 2019], highly sparse interaction records [Chen et al. 2020b], and limited computing resources [Long et al. 2022; Imran et al. 2022]. Secondly, in a broader sense, a trustworthy recommendation/search approach should also be robust, interpretable, secure, privacy-preserving, and fair across different use cases. Specifically, robustness evaluates a model’s performance consistency under various operating conditions like noisy data [Zhang et al. 2020]; interpretability and fairness respectively evaluate if a model can make its decision processes transparent [Chen et al. 2020c;, 2021; Lyu et al. 2021; Cui et al. 2022; Ren et al. 2021] and the decision outcomes unbiased [Chen et al. 2020a; Li et al. 2021; Yin et al. 2012]; while security and privacy respectively emphasize a model’s ability to handle cyber-attacks [Zhang et al. 2021b;, 2022] and to prevent personal information leakage [Zhang and Yin 2022; Zhang et al. 2021c;, 2021a; Yuan et al. 2023; Wang et al. 2022b]. Consequently, trustworthiness is becoming a key performance indicator for state-of-the-art recommendation and search approaches. In light of these emerging challenges, this special section focuses on novel research in this field with the notion of trustworthiness. The articles presented in this special issue will further promote responsible AI applications, thus better universalizing the advanced techniques to a wider range of the common public.

introduction Share on Trustworthy Recommendation and Search: Introduction to the Special Section - Part 2 Authors: Hongzhi Yin The University of Queensland Australia The University of Queensland Australia 0000-0003-1395-261XSearch about this author , Yizhou Sun University of California, Los Angeles, USA University of California, Los Angeles, USA 0000-0003-1812-6843Search about this author , Guandong Xu University of Technology Sydney, Australia University of Technology Sydney, Australia 0000-0003-4493-6663Search about this author , Evangelos Kanoulas University of Amsterdam, Netherlands University of Amsterdam, Netherlands 0000-0002-8312-0694Search about this author Authors Info & Claims ACM Transactions on Information SystemsVolume 41Issue 4Article No.: 82pp 1–6https://doi.org/10.1145/3604776Published:28 July 2023Publication History 0citation0DownloadsMetricsTotal Citations0Total Downloads0Last 12 Months0Last 6 weeks0 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 SiteGet Access

Predicting future events in dynamic knowledge graphs has attracted significant attention. Existing work models the historical information in a holistic way, which achieves satisfactory performance. However, in real-world scenarios, the influence of historical information on future events is changing over time. Therefore, it is difficult to distinguish the historical information of different roles by invariably embedding historical entities with simple vector stacking. Furthermore, it is laborious to explicitly learn a distributed representation of each historical repetitive fact at different timestamps. This poses a challenge to the widely adopted codec-based architectures. In this paper, we propose a novel model for predicting future events, namely Distributed Attention Network (DA-Net). Rather than obtaining the fixed representations of historical events, DA-Net attempts to learn the distributed attention of future events on repetitive facts at different historical timestamps inspired by human cognitive theory. In human cognitive theory, when humans make a decision, similar historical events are replayed during memory recall. Based on memory, the original intention is adjusted according to their recent knowledge developments, making the action more reasonable to the context. Experiments on four benchmark datasets demonstrate a substantial improvement of DA-Net on multiple evaluation metrics.

Temporal knowledge graph (TKG) reasoning has attracted significant attention. Recent approaches for modeling historical information have led to great advances. However, the problems of time variability and unseen entities have become two major obstacles preventing further development. The time variability problem means that different historical timestamps play different roles in the inference process. Furthermore, in the context of time variability, the unseen entity problem means that a query cannot obtain a predicted entity that is unseen in the scale-varying history rather than in a fixed set, thus turning from static to dynamic. In this paper, we propose a novel method named DHU-NET for addressing the time variability challenge and the dynamic unseen entity challenge derived from it. With regard to the former concern, we propose a time-distributed representation learning method based on a graph convolutional network(GCN) and a self-attention mechanism, which learns the distributed representations of facts at different historical timestamps and comprehensively pays different levels of attention to the different timestamps. With regard to the latter issue, we extract the unseen entities from a global static KG based on a copy mechanism and bring them into consideration during the final prediction step. Experiments on six benchmark datasets demonstrate the substantial improvements achieved by DHUNET in terms of multiple evaluation metrics. Our released codes are available at https://github.com/CGCL-codes/DHUNET.

Session-based recommender systems (SBR) are becoming increasingly popular because they can predict user interests without relying on long-term user profile and support login-free recommendation. Modern recommender systems operate in a fully server-based fashion. To cater to millions of users, the frequent model maintaining and the high-speed processing for concurrent user requests are required, which comes at the cost of a huge carbon footprint. Meanwhile, users need to upload their behavior data even including the immediate environmental context to the server, raising the public concern about privacy. On-device recommender systems circumvent these two issues with cost-conscious settings and local inference. However, due to the limited memory and computing resources, on-device recommender systems are confronted with two fundamental challenges: (1) how to reduce the size of regular models to fit edge devices? (2) how to retain the original capacity?

Off-policy learning has drawn huge attention in recommender systems (RS), which provides an opportunity for reinforcement learning to abandon the expensive online training. However, off-policy learning from logged data suffers biases caused by the policy shift between the target policy and the logging policy. Consequently, most off-policy learning resorts to inverse propensity scoring (IPS) which however tends to be over-fitted over exposed (or recommended) items and thus fails to explore unexposed items.

Collaborative filtering is a central task in a broad range of recommender systems. As traditional methods train latent variables for user/item individuals under a transductive setting, it requires re-training for out-of-sample inferences. Inductive matrix completion (IMC) solves this problem by learning transformation functions upon engineered features, but it sacrifices model expressiveness and highly depends on feature qualities. In this paper, we propose Geometric Inductive Matrix Completion (GIMC) by introducing hyperbolic geometry and a unified message passing scheme into this generic task. The proposed method is the earliest attempt utilizing capacious hyperbolic space to enhance the capacity of IMC. It is the first work defining continuous explicit feedback prediction within non-Euclidean space by introducing hyperbolic regression for vertex interactions. This is also the first to provide comprehensive evidence that edge semantics can significantly improve recommendations, which is ignored by previous works. The proposed method outperforms the state-of-the-art algorithms with less than 1% parameters compared to its transductive counterparts. Extensive analysis and ablation studies are conducted to reveal the design considerations and practicability for a positive impact to the research community.

Reinforcement learning has recently become an active topic in recommender system research, where the logged data that records interactions between items and users feedback is used to discover the policy. Much off-policy learning, referring to the procedure of policy optimization with access only to logged feedback data, has been a popular research topic in reinforcement learning. However, the log entries are biased in that the logs over-represent actions favored by the recommender system, as the user feedback contains only partial information limited to the particular items exposed to the user. As a result, the policy learned from such off-line logged data tends to be biased from the true behaviour policy.

Unsupervised graph representation learning has emerged as a powerful tool to address real-world problems and achieves huge success in the graph learning domain. Graph contrastive learning is one of the unsupervised graph representation learning methods, which recently attracts attention from researchers and has achieved state-of-the-art performances on various tasks. The key to the success of graph contrastive learning is to construct proper contrasting pairs to acquire the underlying structural semantics of the graph. However, this key part is not fully explored currently, most of the ways generating contrasting pairs focus on augmenting or perturbating graph structures to obtain different views of the input graph. But such strategies could degrade the performances via adding noise into the graph, which may narrow down the field of the applications of graph contrastive learning. In this paper, we propose a novel graph contrastive learning method, namely Dual Space Graph Contrastive (DSGC) Learning, to conduct graph contrastive learning among views generated in different spaces including the hyperbolic space and the Euclidean space. Since both spaces have their own advantages to represent graph data in the embedding spaces, we hope to utilize graph contrastive learning to bridge the spaces and leverage advantages from both sides. The comparison experiment results show that DSGC achieves competitive or better performances among all the datasets. In addition, we conduct extensive experiments to analyze the impact of different graph encoders on DSGC, giving insights about how to better leverage the advantages of contrastive learning between different spaces.

Graph contrastive learning is the state-of-the-art unsupervised graph representation learning framework and has shown comparable performance with supervised approaches. However, evaluating whether the graph contrastive learning is robust to adversarial attacks is still an open problem because most existing graph adversarial attacks are supervised models, which means they heavily rely on labels and can only be used to evaluate the graph contrastive learning in a specific scenario. For unsupervised graph representation methods such as graph contrastive learning, it is difficult to acquire labels in real-world scenarios, making traditional supervised graph attack methods difficult to be applied to test their robustness. In this paper, we propose a novel unsupervised gradient-based adversarial attack that does not rely on labels for graph contrastive learning. We compute the gradients of the adjacency matrices of the two views and flip the edges with gradient ascent to maximize the contrastive loss. In this way, we can fully use multiple views generated by the graph contrastive learning models and pick the most informative edges without knowing their labels, and therefore can promisingly support our model adapted to more kinds of downstream tasks. Extensive experiments show that our attack outperforms unsupervised baseline attacks and has comparable performance with supervised attacks in multiple downstream tasks including node classification and link prediction. We further show that our attack can be transferred to other graph representation models as well.

The modern industrial sector generates enormous amounts of high-dimensional heterogeneous data daily. However, mostly the vectored data (rank-one tensor) have been considered for anomaly detection, whereas the data in real-life is high dimensional. The expressive power of methods based on vector data is restrictive as they may destroy the structural information embedded in data and lead to the curse-of-dimensionality and overfitting. In this paper, we present a novel anomaly detection approach for large-scale tensor data. We first present novel one-class support tensor machines (OCSTM) with bounded loss function. We further extend it by leveraging the randomness to design a scalable approach that can also be used for large-scale anomaly detection. To solve the corresponding optimization of the objective function, we utilize half-quadratic optimization followed by solving it like a traditional OCSTM optimization at each iteration. We demonstrate the proposed randomized OCSTM with bounded hinge loss through experiments on 14 benchmark data sets. Experimental results demonstrate the effectiveness of the proposed approach against anomalies and a significant reduction in the computational complexity.

Abstract Causal inference is capable of estimating the treatment effect (i.e., the causal effect oftreatmenton theoutcome) to benefit the decision making in various domains. One fundamental challenge in this research is that the treatment assignment bias in observational data. To increase the validity of observational studies on causal inference, representation-based methods as the state-of-the-art have demonstrated the superior performance of treatment effect estimation. Most representation-based methods assume all observed covariates are pre-treatment (i.e., not affected by the treatment) and learn a balanced representation from these observed covariates for estimating treatment effect. Unfortunately, this assumption is often too strict a requirement in practice, as some covariates are changed by doing an intervention on treatment (i.e., post-treatment). By contrast, the balanced representation learned from unchanged covariates thus biases the treatment effect estimation. In light of this, we propose a deep treatment-adaptive architecture (DTANet) that can address the post-treatment covariates and provide a unbiased treatment effect estimation. Generally speaking, the contributions of this work are threefold. First, our theoretical results guarantee DTANet can identify treatment effect from observations. Second, we introduce a novel regularization of orthogonality projection to ensure that the learned confounding representation is invariant and not being contaminated by the treatment, meanwhile mediate variable representation is informative and discriminative for predicting the outcome. Finally, we build on the optimal transport and learn a treatment-invariant representation for the unobserved confounders to alleviate the confounding bias.

Hypergraphs have been becoming a popular choice to model complex, non-pairwise, and higher-order interactions for recommender systems. However, compared with traditional graph-based methods, the constructed hypergraphs are usually much sparser, which leads to a dilemma when balancing the benefits of hypergraphs and the modelling difficulty. Moreover, existing sequential hypergraph recommendation overlooks the temporal modelling among user relationships, which neglects rich social signals from the recommendation data. To tackle the above shortcomings of the existing hypergraph-based sequential recommendations, we propose a novel architecture named Hyperbolic Hypergraph representation learning method for Sequential Recommendation (H2SeqRec) with the pre-training phase. Specifically, we design three self-supervised tasks to obtain the pre-training item embeddings to feed or fuse into the following recommendation architecture (with two ways to use the pre-trained embeddings). In the recommendation phase, we learn multi-scale item embeddings via a hierarchical structure to capture multiple time-span information. To alleviate the negative impact of sparse hypergraphs, we utilize a hyperbolic space-based hypergraph convolutional neural network to learn the dynamic item embeddings. Also, we design an item enhancement module to capture dynamic social information at each timestamp to improve effectiveness. Extensive experiments are conducted on two real-world datasets to prove the effectiveness and high performance of the model.

Advertising is critical to many online e-commerce platforms such as e-Bay and Amazon. One of the important signals that these platforms rely upon is the click-through rate (CTR) prediction. The recent popularity of multi-modal sharing platforms such as TikTok has led to an increased interest in online micro-videos. It is, therefore, useful to consider micro-videos to help a merchant target micro-video advertising better and find users' favourites to enhance user experience. Existing works on CTR prediction largely exploit unimodal content to learn item representations. A relatively minimal effort has been made to leverage multi-modal information exchange among users and items. We propose a model to exploit the temporal user-item interactions to guide the representation learning with multi-modal features, and further predict the user click rate of the micro-video item. We design a Hypergraph Click-Through Rate prediction framework (HyperCTR) built upon the hyperedge notion of hypergraph neural networks, which can yield modal-specific representations of users and micro-videos to better capture user preferences. We construct a time-aware user-item bipartite network with multi-modal information and enrich the representation of each user and item with the generated interests-based user hypergraph and item hypergraph. Through extensive experiments on three public datasets, we demonstrate that our proposed model significantly outperforms various state-of-the-art methods.

Hashtag, a product of user tagging behavior, which can well describe the semantics of the user-generated content personally over social network applications, e.g., the recently popular micro-videos. Hashtags have been widely used to facilitate various micro-video retrieval scenarios, such as search engine and categorization. In order to leverage hashtags on micro-media platform for effective e-commerce marketing campaign, there is a demand from e-commerce industry to develop a mapping algorithm bridging its categories and micro-video hashtags. In this demo paper, we therefore proposed a novel solution called TagPick that incorporates clues from all user behavior metadata (hashtags, interactions, multimedia information) as well as relational data (graph-based network) into a unified system to reveal the correlation between e-commerce categories and hashtags in industrial scenarios. In particular, we provide a tag-level popularity strategy to recommend the relevant hashtags for e-Commerce platform (e.g., eBay).

Underwriting is an important process in insurance and is concerned with accepting individuals into insurance policy with tolerable claim risk. Underwriting is a tedious and labor intensive process relying on underwriters' domain knowledge and experience, thus is labor intensive and prone to error. Machine learning models are recently applied to automate the underwriting process and thus to ease the burden on the underwriters as well as improve underwriting accuracy. However, observational data used for underwriting modelling is high dimensional, sparse and incomplete, due to the dynamic evolving nature (e.g., upgrade) of business information systems. Simply applying traditional supervised learning methods e.g., logistic regression or Gradient boosting on such highly incomplete data usually leads to the unsatisfactory underwriting result, thus requiring practical data imputation for training quality improvement. In this paper, rather than choosing off-the-shelf solutions tackling the complex data missing problem, we propose an innovative Generative Adversarial Nets (GAN) framework that can capture the missing pattern from a causal perspective. Specifically, we design a structural causal model to learn the causal relations underlying the missing pattern of data. Then, we devise a Causality-aware Generative network (CaGen) using the learned causal relationship prior to generating missing values, and correct the imputed values via the adversarial learning. We also show that CaGen significantly improves the underwriting prediction in real-world insurance applications.

User purchasing prediction with multi-behavior information remains a challenging problem for current recommendation systems. Various methods have been proposed to address it via leveraging the advantages of graph neural networks (GNNs) or multi-task learning. However, most existing works do not take the complex dependencies among different behaviors of users into consideration. They utilize simple and fixed schemes, like neighborhood information aggregation or mathematical calculation of vectors, to fuse the embeddings of different user behaviors to obtain a unified embedding to represent a user’s behavioral patterns which will be used in downstream recommendation tasks. To tackle the challenge, in this paper, we first propose the concept of hyper meta-path to construct hyper meta-paths or hyper meta-graphs to explicitly illustrate the dependencies among different behaviors of a user. How to obtain a unified embedding for a user from hyper meta-paths and avoid the previously mentioned limitations simultaneously is critical. Thanks to the recent success of graph contrastive learning, we leverage it to learn embeddings of user behavior patterns adaptively instead of assigning a fixed scheme to understand the dependencies among different behaviors. A new graph contrastive learning based framework is proposed by coupling with hyper meta-paths, namely HMG-CR, which consistently and significantly outperforms all baselines in extensive comparison experiments.

Hyperbolic space and hyperbolic embeddings are becoming a popular research field for recommender systems. However, it is not clear under what circumstances the hyperbolic space should be considered. To fill this gap, This paper provides theoretical analysis and empirical results on when and where to use hyperbolic space and hyperbolic embeddings in recommender systems. Specifically, we answer the questions that which type of models and datasets are more suited for hyperbolic space, as well as which latent size to choose. We evaluate our answers by comparing the performance of Euclidean space and hyperbolic space on different latent space models in both general item recommendation domain and social recommendation domain, with 6 widely used datasets and different latent sizes. Additionally, we propose a new metric learning based recommendation method called SCML and its hyperbolic version HSCML. We evaluate our conclusions regarding hyperbolic space on SCML and show the state-of-the-art performance of hyperbolic space by comparing HSCML with other baseline methods.

At present, most research on the fairness of recommender systems is conducted either from the perspective of customers or from the perspective of product (or service) providers. However, such a practice ignores the fact that when fairness is guaranteed to one side, the fairness and rights of the other side are likely to reduce. In this paper, we consider recommendation scenarios from the perspective of two sides (customers and providers). From the perspective of providers, we consider the fairness of the providers' exposure in recommender system. For customers, we consider the fairness of the reduced quality of recommendation results due to the introduction of fairness measures. We theoretically analyzed the relationship between recommendation quality, customers fairness, and provider fairness, and design a two-sided fairness-aware recommendation model (TFROM) for both customers and providers. Specifically, we design two versions of TFROM for offline and online recommendation. The effectiveness of the model is verified on three real-world data sets. The experimental results show that TFROM provides better two-sided fairness while still maintaining a higher level of personalization than the baseline algorithms.

Twitter is currently a popular online social media platform which allows users to share their user-generated content. This publicly-generated user data is also crucial to healthcare technologies because the discovered patterns would hugely benefit them in several ways. One of the applications is in automatically discovering mental health problems, e.g., depression. Previous studies to automatically detect a depressed user on online social media have largely relied upon the user behaviour and their linguistic patterns including user's social interactions. The downside is that these models are trained on several irrelevant content which might not be crucial towards detecting a depressed user. Besides, these content have a negative impact on the overall efficiency and effectiveness of the model. To overcome the shortcomings in the existing automatic depression detection methods, we propose a novel computational framework for automatic depression detection that initially selects relevant content through a hybrid extractive and abstractive summarization strategy on the sequence of all user tweets leading to a more fine-grained and relevant content. The content then goes to our novel deep learning framework comprising of a unified learning machinery comprising of Convolutional Neural Network (CNN) coupled with attention-enhanced Gated Recurrent Units (GRU) models leading to better empirical performance than existing strong baselines.

Recent advances in path-based explainable recommendation systems have attracted increasing attention thanks to the rich information provided by knowledge graphs. Most existing explainable recommendation only utilizes static knowledge graph and ignores the dynamic user-item evolutions, leading to less convincing and inaccurate explanations. Although there are some works that realize that modelling user's temporal sequential behaviour could boost the performance and explainability of the recommender systems, most of them either only focus on modelling user's sequential interactions within a path or independently and separately of the recommendation mechanism. In this paper, we propose a novel Temporal Meta-path Guided Explainable Recommendation (TMER), which utilizes well-designed item-item path modelling between consecutive items with attention mechanisms to sequentially model dynamic user-item evolutions on dynamic knowledge graph for explainable recommendations. Compared with existing works that use heavy recurrent neural networks to model temporal information, we propose simple but effective neural networks to capture users' historical item features and path-based context to characterise next purchased item. Extensive evaluations of TMER on three real-world benchmark datasets show state-of-the-art performance compared against recent strong baselines.

Harnessing data from social media to monitor health events is a promising avenue for public health surveillance. A key step is the detection of reports of a disease (referred to as ‘health mention classification’) amongst tweets that mention disease words. Prior work shows that figurative usage of disease words may prove to be challenging for health mention classification. Since the experience of a disease is associated with a negative sentiment, we present a method that utilises sentiment information to improve health mention classification. Specifically, our classifier for health mention classification combines pre-trained contextual word representations with sentiment distributions of words in the tweet. For our experiments, we extend a benchmark dataset of tweets for health mention classification, adding over 14k manually annotated tweets across diseases. We also additionally annotate each tweet with a label that indicates if the disease words are used in a figurative sense. Our classifier outperforms current SOTA approaches in detecting both health-related and figurative tweets that mention disease words. We also show that tweets containing disease words are mentioned figuratively more often than in a health-related context, proving to be challenging for classifiers targeting health-related tweets.

Recent decades have witnessed the rapid growth of educational data mining (EDM), which aims at automatically extracting valuable information from large repositories of data generated by or related to people’s learning activities in educational settings. One of the key EDM tasks is cognitive modelling with examination data, and cognitive modelling tries to profile examinees by discovering their latent knowledge state and cognitive level (e.g. the proficiency of specific skills). However, to the best of our knowledge, the problem of extracting information from both objective and subjective examination problems to achieve more precise and interpretable cognitive analysis remains underexplored. To this end, we propose a fuzzy cognitive diagnosis framework (FuzzyCDF) for examinees’ cognitive modelling with both objective and subjective problems. Specifically, to handle the partially correct responses on subjective problems, we first fuzzify the skill proficiency of examinees. Then we combine fuzzy set theory and educational hypotheses to model the examinees’ mastery on the problems based on their skill proficiency. Finally, we simulate the generation of examination score on each problem by considering slip and guess factors. In this way, the whole diagnosis framework is built. For further comprehensive verification, we apply our FuzzyCDF to three classical cognitive assessment tasks, i.e., predicting examinee performance, slip and guess detection, and cognitive diagnosis visualization. Extensive experiments on three real-world datasets for these assessment tasks prove that FuzzyCDF can reveal the knowledge states and cognitive level of the examinees effectively and interpretatively.

Short-head and long-tail distributed data are widely observed in the real world. The same is true of recommender systems (RSs), where a small number of popular items dominate the choices and feedback data while the rest only account for a small amount of feedback. As a result, most RS methods tend to learn user preferences from popular items since they account for most data. However, recent research in e-commerce and marketing has shown that future businesses will obtain greater profit from long-tail selling. Yet, although the number of long-tail items and users is much larger than that of short-head items and users, in reality, the amount of data associated with long-tail items and users is much less. As a result, user preferences tend to be popularity-biased. Furthermore, insufficient data makes long-tail items and users more vulnerable to shilling attack. To improve the quality of recommendations for items and users in the tail of distribution, we propose a coupled regularization approach that consists of two latent factor models: C-HMF, for enhancing credibility, and S-HMF, for emphasizing specialty on user choices. Specifically, the estimates learned from C-HMF and S-HMF recurrently serve as the empirical priors to regularize one another. Such coupled regularization leads to the comprehensive effects of final estimates, which produce more qualitative predictions for both tail users and tail items. To assess the effectiveness of our model, we conduct empirical evaluations on large real-world datasets with various metrics. The results prove that our approach significantly outperforms the compared methods.

Contextual factors can benefit music recommendation and retrieval tasks remarkably. However, how to acquire and utilize the contextual information still need to be studied. In this paper, we propose a context aware music recommendation approach, which can recommend music appropriate for users' contextual preference for music. In analogy to matrix factorization methods for collaborative filtering, the proposed approach does not require songs to be described by features beforehand, but it learns music pieces' embeddings (vectors in low-dimensional continuous space) from music playing records and corresponding metadata and infer users' general and contextual preference for music from their playing records with the learned embedding. Then, our approach can recommend appropriate music pieces. Experimental evaluations on a real world dataset show that the proposed approach outperforms baseline methods.

In the real-world environment, users have sufficient experience in their focused domains but lack experience in other domains. Recommender systems are very helpful for recommending potentially desirable items to users in unfamiliar domains, and cross-domain collaborative filtering is therefore an important emerging research topic. However, it is inevitable that the cold-start issue will be encountered in unfamiliar domains due to the lack of feedback data. The Bayesian approach shows that priors play an important role when there are insufficient data, which implies that recommendation performance can be significantly improved in cold-start domains if informative priors can be provided. Based on this idea, we propose a Weighted Irregular Tensor Factorization (WITF) model to leverage multi-domain feedback data across all users to learn the cross-domain priors w.r.t. both users and items. The features learned from WITF serve as the informative priors on the latent factors of users and items in terms of weighted matrix factorization models. Moreover, WITF is a unified framework for dealing with both explicit feedback and implicit feedback. To prove the effectiveness of our approach, we studied three typical real-world cases in which a collection of empirical evaluations were conducted on real-world datasets to compare the performance of our model and other state-of-the-art approaches. The results show the superiority of our model over comparison models.

With the pervasive use of GPS-enabled smart phones, location-based services, e.g., Location Based Social Networking (LBSN) have emerged . Point-of-Interests (POIs) Recommendation, as a typical component in LBSN, provides additional values to both customers and merchants in terms of user experience and business turnover. Existing POI recommendation systems mainly adopt Collaborative Filtering (CF), which only exploits user given ratings (i.e., user overall evaluation) about a merchant while regardless of the user preference difference across multiple aspects, which exists commonly in real scenarios. Meanwhile, besides ratings, most LBSNs also provide the review function to allow customers to give their opinions when dealing with merchants, which is often overlooked in these recommender systems. In this demo, we present MARS, a novel POI recommender system based on multi-aspect user preference learning from reviews by using utility theory. We first introduce the organization of our system, and then show how the user preferences across multiple aspects are integrated into our system alongside several case studies of mining user preference and POI recommendations.

With the pervasive use of mobile devices, Location Based Social Networks(LBSNs) have emerged in past years. These LBSNs, allowing their users to share personal experiences and opinions on visited merchants, have very rich and useful information which enables a new breed of location-based services, namely, Merchant Recommendation. Existing techniques for merchant recommendation simply treat each merchant as an item and apply conventional recommendation algorithms, e.g., Collaborative Filtering, to recommend merchants to a target user. However, they do not differentiate the user's real preferences on various aspects, and thus can only achieve limited success. In this paper, we aim to address this problem by utilizing and analyzing user reviews to discover user preferences in different aspects. Following the intuition that a user rating represents a personalized rational choice, we propose a novel utility-based approach by combining collaborative and individual views to estimate user preference (i.e., rating). An optimization algorithm based on a Gaussian model is developed to train our merchant recommendation approach. Lastly we evaluate the proposed approach in terms of effectiveness, efficiency and cold-start using two real-world datasets. The experimental results show that our approach outperforms the state-of-the-art methods. Meanwhile, a real mobile application is implemented to demonstrate the practicability of our method.

Inspired by the actual trend of empowering traditional Web search methodologies by means of novel computational paradigms, in this paper we propose and experimentally assess WebClustCube, a novel system that allows OLAP-enabled Web search of complex objects, thus adding new value to the potentialities of current Web search paradigms. In particular, WebClustCube supports the building and the interactive manipulation of OLAP-enabled Web views over complex objects extracted from distributed databases. The data management, OLAP-like support of WebClustCube is provided by ClustCube, a state-of-the-art framework for coupling OLAP methodologies and clustering algorithms with the goal of analyzing and mining of complex database objects. A case study that clearly shows the potentialities of WebClustCube in the context of next-generation Web search environments is provided. We complement of analytical contribution by means of an experimental assessment and analysis of WebClustCube according to several metric perspectives.

Nowadays, a huge amount of text is being generated for social networking purpose on the Web. Keyword extraction from such text benefit many applications such as advertising, search, and content filtering. Recent studies show that graph based ranking is more effective than traditional term or document frequecy based approaches. However, most work in the literature constructs word to word graph within a document or a collection of documents before applying a kind of random walk. Such a graph does not consider the influence of document importance on keyword extraction. Moreover, social text like a microblog post usually has speical social features such as hashtag and so on, which can help us understand its topic. In this paper, we propose hashtag biased ranking for keyword extraction from a collection of microblog posts. We first build a word-post weighted graph by taking into account the posts themselves. Then, a hashtag biased random walk is applied on this graph, which guides our approach to extract keywords according to the hashtag topic. Last, the final ranking of a word is determined by the stationary probability after a number of interations. We evaluate our proposed method on a real Chinese microblog posts. Experiments show that our method is more effective than the traditional word to word graph based ranking in terms of precision.

The backbone is the natural abstraction of a complex network, which can help people to understand it in a more simplified form. Backbone extraction becomes more challenging as many networks are evolving into large scale and the weight distributions are spanning several orders of magnitude. Traditional filter-based methods tend to include many outliers into the backbone. What is more, they often suffer from the computational inefficiency-the exhaustive search of all nodes or edges is often prohibitively expensive. In this work, we propose a Local Search based Backbone Extraction Heuristic (LS-BEH) to find the backbone in a complex weighted network. First, a strict filtering rule is carefully designed to determine edges to be preserved or discarded. Second, we present a local search model to examine part of edges in an iterative way. Experimental results on two real-life networks demonstrate the advantage of LS-BEH over the classic disparity filter method by either effectiveness or efficiency validity.

Entity resolution identifies entities from different data sources that refer to the same real-world entity and it is an important prerequisite for integrating data from multiple sources. Entity resolution mainly relies on similarity measures on data records. Unfortunately, the data quality of data sources is not so good in practice. Especially web data sources often only provide incomplete information, which leads to the difficulties of direct applying similarity measures to identify the same entities. In order to address this problem, the concept of confidence is introduced to measure the trustworthy of the similarity calculation. An adaptive rule-based approach is used to calculate the similarity between records and its confidence is also derived. Then the similarity and confidence are propagated on the entity relational graph until fix point is reached. Finally, any pair of two records can be determined as matched or unmatched based on a threshold. We performed a series of experiments on real data sets and experiment results show that our approach has a better performance comparing with others.

Choice modeling (CM) aims to describe and predict choices according to attributes of subjects and options. If we presume each choice making as the formation of link between subjects and options, immediately CM can be bridged to link analysis and prediction (LAP) problem. However, such a mapping is often not trivial and straightforward. In LAP problems, the only available observations are links among objects but their attributes are often inaccessible. Therefore, we extend CM into a latent feature space to avoid the need of explicit attributes. Moreover, LAP is usually based on binary linkage assumption that models observed links as positive instances and unobserved links as negative instances. Instead, we use a weaker assumption that treats unobserved links as pseudo negative instances. Furthermore, most subjects or options may be quite heterogeneous due to the long-tail distribution, which is failed to capture by conventional LAP approaches. To address above challenges, we propose a Bayesian heteroskedastic choice model to represent the non-identically distributed linkages in the LAP problems. Finally, the empirical evaluation on real-world datasets proves the superiority of our approach.

The problem of software artifact retrieval has the goal to effectively locate software artifacts, such as a piece of source code, in a large code repository. This problem has been traditionally addressed through the textual query. In other words, information retrieval techniques will be exploited based on the textual similarity between queries and textual representation of software artifacts, which is generated by collecting words from comments, identifiers, and descriptions of programs. However, in addition to these semantic information, there are rich information embedded in source codes themselves. These source codes, if analyzed properly, can be a rich source for enhancing the efforts of software artifact retrieval. To this end, in this paper, we develop a feature extraction method on source codes. Specifically, this method can capture both the inherent information in the source codes and the semantic information hidden in the comments, descriptions, and identifiers of the source codes. Moreover, we design a heterogeneous metric learning approach, which allows to integrate code features and text features into the same latent semantic space. This, in turn, can help to measure the artifact similarity by exploiting the joint power of both code and text features. Finally, extensive experiments on real-world data show that the proposed method can help to improve the performances of software artifact retrieval with a significant margin.

Exploiting label dependency is a key challenge in multi-label learning, and current methods solve this problem mainly by training models on the combination of related labels and original features. However, label dependency cannot be exploited dynamically and mutually in this way. Therefore, we propose a novel paradigm of leveraging label dependency in an iterative way. Specifically, each label's prediction will be updated and also propagated to other labels via an random walk with restart process. Meanwhile, the label propagation is implemented as a supervised learning procedure via optimizing a loss function, thus more appropriate label dependency can be learned. Extensive experiments are conducted, and the results demonstrate that our method can achieve considerable improvements in terms of several evaluation metrics.

Collaborative filtering (CF) is a major technique in recommender systems to help users find their potentially desired items. Since the data sparsity problem is quite commonly encountered in real-world scenarios, Cross-Domain Collaborative Filtering (CDCF) hence is becoming an emerging research topic in recent years. However, due to the lack of sufficient dense explicit feedbacks and even no feedback available in users' uninvolved domains, current CDCF approaches may not perform satisfactorily in user preference prediction. In this paper, we propose a generalized Cross Domain Triadic Factorization (CDTF) model over the triadic relation user-item-domain, which can better capture the interactions between domain-specific user factors and item factors. In particular, we devise two CDTF algorithms to leverage user explicit and implicit feedbacks respectively, along with a genetic algorithm based weight parameters tuning algorithm to trade off influence among domains optimally. Finally, we conduct experiments to evaluate our models and compare with other state-of-the-art models by using two real world datasets. The results show the superiority of our models against other comparative models.

As a prevalent type of Web advertising, contextual advertising refers to the placement of the most relevant ads into a Web page, so as to increase the number of ad-clicks. However, some problems of homonymy and polysemy, low intersection of keywords etc., can lead to the selection of irrelevant ads for a page. In this paper, we present a new contextual advertising approach to overcome the problems, which uses Wikipedia concept and category information to enrich the content representation of an ad (or a page). First, we map each ad and page into a keyword vector, a concept vector and a category vector. Next, we select the relevant ads for a given page based on a similarity metric that combines the above three feature vectors together. Last, we evaluate our approach by using real ads, pages, as well as a great number of concepts and categories of Wikipedia. Experimental results show that our approach can improve the precision of ads-selection effectively.

Due to the inherent correlation among Web objects and the lack of a uniform schema of Web documents, Web community mining and analysis has become an important area for Web data management and analysis. The research of Web communities spans a number of research domains such as Web mining, Web search, clustering and text retrieval. In this talk we will present some recent studies on this topic, which cover finding relevant Web pages based on linkage information, discovering user access patterns through analyzing Web log files, co-clustering Web objects and investigating social networks from Web data. The algorithmic issues and related experimental studies will be addressed. Some research directions are also to be discussed.