Large language models (LLMs) are increasingly used for recommendation reranking, but their listwise predictions can depend on the order in which candidates are presented. This creates a mismatch between the set-based nature of recommendation and the sequence-based computation of decoder-only LLMs, where permuting an otherwise identical candidate set can change item scores and final rankings. Such order sensitivity makes LLM-based rerankers difficult to rely on, since rankings may reflect prompt serialization rather than user preference. We propose InvariRank, a permutation-invariant listwise reranking framework that addresses this dependence at the architectural level. InvariRank blocks cross-candidate attention with a structured attention mask and negates position-induced scoring changes through shared positional framing under Rotary Positional Embeddings (RoPE). Combined with a listwise learning-to-rank objective, the model scores all candidates in a single forward pass, avoiding permutation-based invariance training objectives that require multiple permutations of a candidate set. Experiments on recommendation benchmarks show that InvariRank maintains competitive ranking effectiveness while producing stable rankings across candidate permutations. The results suggest that architectural invariance is a practical route to reliable and efficient LLM-based recommendation reranking. The source code is at https://github.com/ejbito/InvariRank.

Large-language-model (LLM) based user simulation is increasingly adopted for evaluating search engines, recommender systems, and retrieval-augmented generation pipelines, yet most simulators remain opaque: it is difficult to determine why a simulated user made a particular choice or whether that choice is consistent with the intended user profile. Compounding this, recent research shows that LLMs can produce biased or discriminatory responses depending on user background characteristics such as language, education level, and cultural context, raising concerns about the equitable treatment of minority and disadvantaged groups. This half-day, in-person tutorial introduces a proposed design-and-audit framework that treats a user simulator as a verifiable engineering artefact composed of seven auditable components---structured Persona, task-aware Contract, matched human-vs-agent Execution, auditable Trace, persona-aligned Verification, structured Feedback, and a Refinement loop that updates personas and contracts. Through two hands-on mini-labs on recommendation-list evaluation and search-query formulation, participants will inspect simulator behaviour end-to-end, distinguish diagnostic discrepancy analysis from statistical validation, and apply checks for fidelity, credibility, and demographic bias. The tutorial targets information retrieval and recommender systems researchers and practitioners interested in user behaviour simulation and responsible AI.

Large language models (LLMs) are widely used for question answering (QA) but can generate biased or stereotype-driven answers due to demographic associations learned during pre-training. Existing mitigation strategies often rely on model access or fine-tuning, which limits their applicability to closed-source LLMs. We propose a Causal Front Door Prompting framework (CFDP) that reduces demographic influence by intervening on the chain of thought reasoning, which is treated as an observable mediator. CFDP samples and clusters multiple reasoning traces and estimates answer probabilities through weighted aggregation. Experiments on two widely used bias-sensitive QA benchmarks, BBQ and Stereotype, across major LLMs show that CFDP consistently improves fairness metrics without sacrificing QA accuracy. Ablation and sensitivity analyses confirm the value of each component, indicating that causal intervention on reasoning provides an effective and practical approach for bias mitigation in LLM-based QA.

Recommender systems play a critical role in digital platforms by providing personalized suggestions based on user preferences and interactions. While Graph Neural Networks (GNNs) have improved recommendation accuracy by modeling complex user-item relationships, existing systems often struggle to integrate heterogeneous data sources, including textual reviews, item attributes, and user profiles, into a unified and explainable framework. To address these limitations, we present ExpLMGCN, an explainable multi-modal recommender system that fuses user-item interactions, review embeddings, item features, and user profiles using GNNs and pre-trained language models. Fine-grained user preferences are captured by extracting opinion-aspect pairs (OAs) from reviews and representing them as User–OA and Item–OA bipartite graphs. A contrastive learning framework aligns multi-modal embeddings into a shared latent space, while explanations are generated by ranking OAs based on preference similarity, sentiment, novelty, and factuality. Multiple candidate explanations are produced via large language models, and the optimal one is selected using semantic alignment, multi-aspect coverage, and a self-consistency feedback loop. Extensive experiments on benchmark datasets demonstrate that ExpLMGCN consistently outperforms state-of-the-art baselines, achieving up to 4.39% improvement in recommendation accuracy across Recall and NDCG metrics. Moreover, ExpLMGCN substantially enhances explanation quality, yielding notable gains across BERTScore, ROUGE, and SBERT, and achieving up to 29.4% improvement in BLEU score. These results highlight the model’s ability to provide not only more accurate recommendations but also more coherent and semantically aligned explanations.

Large Language Models (LLMs) have shown impressive capabilities in natural language processing but still struggle to perform well on knowledge-intensive tasks that require deep reasoning and the integration of external knowledge. Although methods such as Retrieval-Augmented Generation (RAG) and Chain-of-Thought (CoT) have been proposed to enhance LLMs with external knowledge, they still suffer from internal bias in LLMs, which often leads to incorrect answers. In this paper, we propose a novel causal prompting framework, Conditional Front-Door Prompting (CFD-Prompting), which enables the unbiased estimation of the causal effect between the query and the answer, conditional on external knowledge, while mitigating internal bias. By constructing counterfactual external knowledge, our framework simulates how the query behaves under varying contexts, addressing the challenge that the query is fixed and is not amenable to direct causal intervention. Compared to the standard front-door adjustment, the conditional variant operates under weaker assumptions, enhancing both robustness and generalisability of the reasoning process. Extensive experiments across multiple LLMs and benchmark datasets demonstrate that CFD-Prompting significantly outperforms existing baselines in both accuracy and robustness.

Marked Temporal Point Process (MTPP) has been well studied to model the event distribution in marked event streams, which can be used to predict the mark and arrival time of the next event. However, existing studies overlook that the distribution of event marks is highly imbalanced in many real-world applications, with some marks being frequent but others rare. The imbalance poses a significant challenge to the performance of the next event prediction, especially for events of rare marks. To address this issue, we propose a thresholding method, which learns thresholds to tune the mark probability normalized by the mark's prior probability to optimize mark prediction, rather than predicting the mark directly based on the mark probability as in existing studies. In conjunction with this method, we predict the mark first and then the time. In particular, we develop a novel neural Marked Temporal Point Process (MTPP) model to support effective time sampling and estimation of mark probability without computationally expensive numerical improper integration. Extensive experiments on real-world datasets demonstrate the superior performance of our solution against various baselines for the next event mark and time prediction. The code is available at https://github.com/undes1red/IFNMTPP.

Traditional offline evaluation methods for recommender systems struggle to capture the complexity of modern platforms due to sparse behavioural signals, noisy data, and limited modelling of user personality traits.While simulation frameworks can generate synthetic data to address these gaps, existing methods fail to replicate behavioural diversity, limiting their effectiveness.To overcome these challenges, we propose the Personality-driven User Behaviour Simulator (PUB), an LLM-based simulation framework that integrates the Big Five personality traits to model personalised user behaviour.PUB dynamically infers user personality from behavioural logs (e.g., ratings, reviews) and item metadata, then generates synthetic interactions that preserve statistical fidelity to real-world data.Experiments on the Amazon review datasets show that logs generated by PUB closely align with real user behaviour and reveal meaningful associations between personality traits and recommendation outcomes.These results highlight the potential of the personality-driven simulator to advance recommender system evaluation, offering scalable, controllable, high-fidelity alternatives to resource-intensive real-world experiments.1

Session-based Recommender Systems (SBRSs) aim at timely predicting the next likely item by capturing users’ current preferences in sessions. Existing SBRSs research only focuses on maximizing session utilities, and little has been done on the fairness issue in SBRSs, which is vital but different from the same issue in traditional Recommender Systems (RSs). To fill in this gap, we define a novel concept ofsession-oriented fairnessto enforce individual items to have the same exposure accumulated within each single session, which is flexible enough to provide opportunities to achieve different fairness goals. Then, we devise a Session-Oriented Fairness-Aware algorithm (SOFA) with a dual Temporal Convolutional Networks (TCN) architecture: one is SOUP (Session-Oriented Utility Promoter) and the other is SODA (Session-Oriented Disparity Alleviator). Benefit from the collaborative learning of SOUP and SODA for the evolution of accumulated exposure in sessions,SOFAis effective to maximize session-oriented fairness while maintaining high session utilities. To the best of our knowledge, this research is the first to solve fairness issues in SBRSs. Extensive experiments on real-world datasets demonstrate thatSOFAoutperforms the state-of-the-art approaches in terms of both utility and fairness.

Recommender systems play a crucial role in personalizing user experiences, yet ensuring fairness in their outcomes remains an elusive challenge. This work explores the impact of individual users or items on the fairness of recommender systems, thus addressing a significant knowledge gap in the field. We introduce an innovative approach called Adding-Based Counterfactual Fairness Reasoning ( ACFR ), designed to elucidate recommendation fairness from the unique perspectives of users and items. Conventional methodologies, like erasing-based counterfactual analysis, pose limitations, particularly in modern recommender systems dealing with a large number of users and items. These traditional methods, by excluding specific users or items, risk disrupting the crucial relational structure central to collaborative filtering recommendations. In contrast, ACFR employs an adding-based counterfactual analysis, a unique strategy allowing us to consider potential, yet-to-happen user-item interactions. This strategy preserves the core user-item relational structure, while predicting future behaviors of users or items. The commonly used feature-based counterfactual analysis, relying on gradient-based optimization to identify interference on each feature, is not directly applicable in our case. In the recommendation scenario we consider, only interactions between users and items are present during model training—no distinct features are involved. Consequently, the traditional mechanism proves impractical for identifying interference on these existing interactions. Our extensive experiments validate the superiority of ACFR over traditional baseline methods, demonstrating significant improvements in recommendation fairness on benchmark datasets. This work, therefore, provides a fresh perspective and a promising methodology for enhancing fairness in recommender systems.

In E-Commence applications, online shoppers increasingly rely on reviews to find answers for their queries on products. Recently, there has been significant research on product question answering using reviews. While many neural models have been proposed to generate answers from reviews to answer product questions, the evaluation of review-based product question answering (RPQA) models is still largely using lexicon-based text similarity metrics against reference answers. In this paper, we propose a distant supervised contrastive learning network based on modern transformer-based language models by leveraging distant ground-truth community question answers to train a model for RPQA evaluation. At deployment, the learned metric does not require reference answers to evaluate the generated answers for product queries. We conducted extensive experiments on publicly available AmazonQA datasets to evaluate RPQA models. Experiment results show that our learned metric correlates well with human judgements, and outperforms existing lexicon-based or embedding-based metrics.

Many existing recommender systems (RSs) assume user behavior is governed solely by their interests. However, the peer effect often influences individual decision-making, which leads to conformity behavior. Conventional solutions that eliminate indiscriminately such bias may cause RSs to neglect valuable information and depersonalize the recommendation results. Also, conformity can transform into user interest, e.g., discovering new tastes after a glance at popular music. By better representing different forms of conformity influence, we can do a better job at interest mining and debiasing. In certain extreme circumstances, the herd effect may be exacerbated by user anxiety with uncertainty (e.g., panic buying during the COVID-19 pandemic). RSs may thus fail to respond in time due to sudden and dramatic changes. Moreover, many existing studies potentially conflate conformity bias with popularity bias and lump together various factors responsible for differences in popularity. In this paper, we identify two distinct types of conformity behavior: informational conformity and normative conformity. To address this, we introduce the TCHN model, which utilizes attentional Hawkes processes to disentangle user self-interest and conformity in a personalized manner. Our approach incorporates temporal graph attention networks to capture users' stable and volatile dynamics. We conduct experiments on three real-world datasets, which uncover diverse levels of conformity among users. The results show that TCHN excels in recommendation accuracy, diversity, and fairness across various user groups.

Sequential recommendations have received great attention in recent years due to their wide application in e-commerce, trip planning, and online education. Contexts reveal the intention of a user in a transaction such as consuming or purchasing an item, and highly affect her behaviour. However, existing techniques cannot fully capture the context-sensitive and dynamic sequential user behaviours for effective and efficient sequential recommendation. To address this limitation, we propose a novel Dynamic Bi-layer Graph learning (DBIG)-based framework for sequential recommendation, which learns the sequence dynamics of user behaviours and contexts. Specifically, we first propose a context graph to identify the relevant contexts to a target user. Then, we propose a transaction context graph that predicts the probability of a transaction happening under certain contexts. In addition, we apply BERT (Bidirectional Encoder Representations from Transformers) to our graph embedding for understanding the user sequential behaviours in bidirectional ways and generating recommendations. Finally, we propose a cluster-structure-based algorithm, which incrementally maintains the DBIG model and BERT over dynamic transaction updates. We evaluate our DBIG-based framework by conducting extensive experiments over five public datasets. The quantitative results demonstrate the superiority of our approach in terms of effectiveness and efficiency.

Human movements in the workspace usually have non-negligible relations with air quality parameters (e.g., CO2, PM2.5, and PM10). We establish a system to monitor indoor human mobility with air quality and assess the interrelationship between these two types of time series data. More specifically, a sensor network was designed in indoor environments to observe air quality parameters continuously. Simultaneously, another sensing module detected participants’ movements around the study areas. In this module, modern data analysis and machine learning techniques have been applied to reconstruct the trajectories of participants with relevant sensor information. Finally, a further study revealed the correlation between human indoor mobility patterns and indoor air quality parameters. Our experimental results demonstrate that human movements in different environments can significantly impact air quality during busy hours. With the results, we propose recommendations for future studies.

We outline a simulation-based study of the effect rapid population-scale concept drifts have on Collaborative Filtering (CF) models. We create a framework for analyzing the effects of macro-trends in population dynamics on the behavior of such models. Our framework characterizes population-scale concept drifts in item preferences and provides a lens to understand the influence events, such as a pandemic, have on CF models. Our experimental results show the initial impact on CF performance at the initial stage of such events, followed by an aggravated population herding effect during the event. The herding introduces a popularity bias that may benefit affected users, but which comes at the expense of a normal user experience. We propose an adaptive ensemble method that can effectively apply optimal algorithms to cope with the change brought about by different stages of the event.

Concept drift in stream data has been well studied in machine learning applications. In the field of recommender systems, this issue is also widely observed, as known as temporal dynamics in user behavior. Furthermore, in the context of COVID-19 pandemic related contingencies, people shift their behavior patterns extremely and tend to imitate others' opinions. The changes in user behavior may not be always rational. Thus, irrational behavior may impair the knowledge learned by the algorithm. It can cause herd effects and aggravate the popularity bias in recommender systems due to the irrational behavior of users. However, related research usually pays attention to the concept drift of individuals and overlooks the synergistic effect among users in the same social group. We conduct a study on user behavior to detect the collaborative concept drifts among users. Also, we empirically study the increase of experience of individuals can weaken herding effects. Our results suggest the CF models are highly impacted by the herd behavior and our findings could provide useful implications for the design of future recommender algorithms.

Existing human mobility forecasting models follow the standard design of the time-series prediction model which takes a series of numerical values as input to generate a numerical value as a prediction. Although treating this as a regression problem seems straightforward, incorporating various contextual information such as the semantic category information of each Place-of-Interest (POI) is a necessary step, and often the bottleneck, in designing an effective mobility prediction model. As opposed to the typical approach, we treat forecasting as a translation problem and propose a novel forecasting through a language generation pipeline. The paper aims to address the human mobility forecasting problem as a language translation task in a sequence-to-sequence manner. A mobility-to-language template is first introduced to describe the numerical mobility data as natural language sentences. The core intuition of the human mobility forecasting translation task is to convert the input mobility description sentences into a future mobility description from which the prediction target can be obtained. Under this pipeline, a two-branch network, SHIFT (Translating Human Mobility Forecasting), is designed. Specifically, it consists of one main branch for language generation and one auxiliary branch to directly learn mobility patterns. During the training, we develop a momentum mode for better connecting and training the two branches. Extensive experiments on three real-world datasets demonstrate that the proposed SHIFT is effective and presents a new revolutionary approach to forecasting human mobility.

Ranking algorithms in recommender systems influence people to make decisions. Conventional ranking algorithms based on implicit feedback data aim to maximize the utility to users by capturing users’ preferences over items. However, these utility-focused algorithms tend to cause fairness issues that require careful consideration in online platforms. Existing fairness-focused studies does not explicitly consider the problem of lacking negative feedback in implicit feedback data, while previous utility-focused methods ignore the importance of fairness in recommendations. To fill this gap, we propose a Generative Adversarial Networks (GANs) based learning algorithm FairGAN mapping the exposure fairness issue to the problem of negative preferences in implicit feedback data. FairGAN does not explicitly treat unobserved interactions as negative, but instead, adopts a novel fairness-aware learning strategy to dynamically generate fairness signals. This optimizes the search direction to make FairGAN capable of searching the space of the optimal ranking that can fairly allocate exposure to individual items while preserving users’ utilities as high as possible.

Factorization models have been successfully applied to the recommendation problems and have significant impact to both academia and industries in the field of Collaborative Filtering ( CF ). However, the intermediate data generated in factorization models’ decision making process (or training process , footprint ) have been overlooked even though they may provide rich information to further improve recommendations. In this article, we introduce the concept of Convergence Pattern, which records how ratings are learned step-by-step in factorization models in the field of CF. We show that the concept of Convergence Patternexists in both the model perspective (e.g., classical Matrix Factorization ( MF ) and deep-learning factorization) and the training (learning) perspective (e.g., stochastic gradient descent ( SGD ), alternating least squares ( ALS ), and Markov Chain Monte Carlo ( MCMC )). By utilizing the Convergence Pattern, we propose a prediction model to estimate the prediction reliability of missing ratings and then improve the quality of recommendations. Two applications have been investigated: (1) how to evaluate the reliability of predicted missing ratings and thus recommend those ratings with high reliability. (2) How to explore the estimated reliability to adjust the predicted ratings to further improve the predication accuracy. Extensive experiments have been conducted on several benchmark datasets on three recommendation tasks: decision-aware recommendation, rating predicted, and Top- N recommendation. The experiment results have verified the effectiveness of the proposed methods in various aspects.

Heterogeneity and irregularity of multi-source data sets present a significant challenge to time-series analysis. In the literature, the fusion of multi-source time-series has been achieved either by using ensemble learning models which ignore temporal patterns and correlation within features or by defining a fixed-size window to select specific parts of the data sets. On the other hand, many studies have shown major improvement to handle the irregularity of time-series, yet none of these studies has been applied to multi-source data. In this work, we design a novel architecture, PIETS, to model heterogeneous time-series. PIETS has the following characteristics: (1) irregularity encoders for multi-source samples that can leverage all available information and accelerate the convergence of the model; (2) parallelised neural networks to enable flexibility and avoid information overwhelming; and (3) attention mechanism that highlights different information and gives high importance to the most related data. Through extensive experiments on real-world data sets related to COVID-19, we show that the proposed architecture is able to effectively model heterogeneous temporal data and outperforms other state-of-the-art approaches in the prediction task.

Human mobility prediction is a core functionality in many location-based services and applications. However, due to the sparsity of mobility data, it is not an easy task to predict future POIs (place-of-interests) that are going to be visited. In this paper, we propose MobTCast, a Transformer-based context-aware network for mobility prediction. Specifically, we explore the influence of four types of context in mobility prediction: temporal, semantic, social, and geographical contexts. We first design a base mobility feature extractor using the Transformer architecture, which takes both the history POI sequence and the semantic information as input. It handles both the temporal and semantic contexts. Based on the base extractor and the social connections of a user, we employ a self-attention module to model the influence of the social context. Furthermore, unlike existing methods, we introduce a location prediction branch in MobTCast as an auxiliary task to model the geographical context and predict the next location. Intuitively, the geographical distance between the location of the predicted POI and the predicted location from the auxiliary branch should be as close as possible. To reflect this relation, we design a consistency loss to further improve the POI prediction performance. In our experimental results, MobTCast outperforms other state-of-the-art next POI prediction methods. Our approach illustrates the value of including different types of context in next POI prediction.

We investigate the impact of popularity bias in false-positive metrics in the offline evaluation of recommender systems. Unlike their true-positive complements, false-positive metrics reward systems that minimize recommendations disliked by users. Our analysis is, to the best of our knowledge, the first to show that false-positive metrics tend to penalise popular items, the opposite behavior of true-positive metrics—causing a disagreement trend between both types of metrics in the presence of popularity biases. We present a theoretical analysis of the metrics that identifies the reason that the metrics disagree and determines rare situations where the metrics might agree—the key to the situation lies in the relationship between popularity and relevance distributions, in terms of their agreement and steepness —two fundamental concepts we formalize. We then examine three well-known datasets using multiple popular true- and false-positive metrics on 16 recommendation algorithms. Specific datasets are chosen to allow us to estimate both biased and unbiased metric values. The results of the empirical study confirm and illustrate our analytical findings. With the conditions of the disagreement of the two types of metrics established, we then determine under which circumstances true-positive or false-positive metrics should be used by researchers of offline evaluation in recommender systems. 1

Network alignment is useful for multiple applications that require increasingly large graphs to be processed. Existing research approaches this as an optimization problem or computes the similarity based on node representations. However, the process of aligning every pair of nodes between relatively large networks is time-consuming and resource-intensive. In this paper, we propose a framework, called G-CREWE (Graph CompREssion With Embedding) to solve the network alignment problem. G-CREWE uses node embeddings to align the networks on two levels of resolution, a fine resolution given by the original network and a coarse resolution given by a compressed version, to achieve an efficient and effective network alignment. The framework first extracts node features and learns the node embedding via a Graph Convolutional Network (GCN). Then, node embedding helps to guide the process of graph compression and finally improve the alignment performance. As part of G-CREWE, we also propose a new compression mechanism called MERGE (Minimum DEgRee NeiGhbors ComprEssion) to reduce the size of the input networks while preserving the consistency in their topological structure. Experiments on all real networks show that our method is more than twice as fast as the most competitive existing methods while maintaining high accuracy.

Fog computing is an emergent computing paradigm that extends the cloud paradigm to the edge. With the explosive growth of smart devices and massive data generated everyday, cloud computing no longer matches the requirements of the Internet of Things (IoT) era, such as low latency, uninterrupted service and location awareness. Thus, fog computing has been introduced as a complement of the current cloud computing model to meet the requirements in IoT. Fog computing is a relatively new networking paradigm and considered as a promising solution to support IoT scenarios. On the one hand, fog computing inherits many features from cloud; on the other hand, fog computing also inherits some challenges and issues from cloud computing: privacy issue is one of them. In this paper, we propose a personalized differential privacy model based on the distance between two fog nodes in a fog network. We also identify the collusion attack in differential privacy framework which compromised the personalized Laplace function. Based on that, we develop a personalized differential privacy model, which not only eliminate this particular attack but also optimize the trade-off between privacy preserving and data utility.

We introduce the novel research problem of task recognition in daily life. We recognize tasks such as project management, planning, meal-breaks, communication, documentation, and family care. We capture Cyber, Physical, and Social (CPS) activities of 17 participants over four weeks using device-based sensing, app activity logging, and an experience sampling methodology. Our cohort includes students, casual workers, and professionals, forming the first real-world context-rich task behaviour dataset. We model CPS activities across different task categories, results highlight the importance of considering the CPS feature sets in modelling, especially work-related tasks.

False-positive metrics can capture an important side of recommendation quality, focusing on the impact of suggestions that are disliked by users, as a complement of common metrics that only measure the amount of successful recommendations. In this paper we research the extent to which false-positive metrics agree or disagree with true-positive metrics in the offline evaluation of recommender systems. We discover a surprising degree of systematic disagreement that was occasionally noted but not explained in the literature by previous authors. We find an explanation for the discrepancy be-tween the metrics in the effect of popularity biases, which impact false and true-positive metrics in very different ways: instead of rewarding the recommendation of popular items, as with true-positive, false-positive metrics penalize the popular. We determine precise conditions and cases in the general trends, with a formal explanation for our findings, which we confirm and illustrate empirically in experiments with different datasets.

One of the core challenges in open-plan workspaces is to ensure a good level of concentration for the workers while performing their tasks. Hence, being able to infer concentration levels of workers will allow building designers, managers, and workers to estimate what effect different open-plan layouts will have and to find an optimal one. In this article, we present an ambient-physical system to investigate the concentration inference problem. Specifically, we deploy a series of pervasive sensors to capture various ambient and physical signals related to perceived concentration at work. The practicality of our system has been tested on two large open-plan workplaces with different designs and layouts. The empirical results highlight promising applications of pervasive sensing in occupational concentration inference, which can be adopted to enhance the capabilities of modern workplaces.

Fog computing is an emergent computing paradigm that extends the cloud paradigm. With the explosive growth of smart devices and mobile users, cloud computing no longer matches the requirements of the Internet of Things (IoT) era. Fog computing is a promising solution to satisfying these new requirements, such as low latency, uninterrupted service, and location awareness. As a typical new computing paradigm and network architecture, fog computing raises new challenges, such as privacy, data management, data analytics, information overload, and participatory sensing. In this article, we present a fog-based hybrid recommender system to address the issue of information overload in fog computing. Our proposed system not only abstracts useful information from the fog environment but can also be considered as an optimization tool due to its ability to provide suggestions to improve system performance. In particular, we demonstrate that the proposed system provides personalized and localized recommendations to users, and also advise the system itself to precache the content to optimize the storage capacity of the fog server.

This article investigates the cyber-physical behavior of users in a large indoor shopping mall by leveraging anonymized (opt in) Wi-Fi association and browsing logs recorded by the mall operators. Our analysis shows that many users exhibit a high correlation between their cyber activities and their physical context. To find this correlation,propose a mechanism to semantically label a physical space with rich categorical information from DBPedia concepts and compute a contextual similarity that represents a user’s activities with the mall context. We demonstrate the application of cyber-physical contextual similarity in two situations: user visit intent classification and future location prediction. The experimental results demonstrate that exploitation of contextual similarity significantly improves the accuracy of such applications.

Intelligent assistants can serve many purposes, including entertainment (e.g. playing music), home automation, and task management (e.g. timers, reminders). The role of these assistants is evolving to also support people engaged in work tasks, in workplaces and beyond. To design truly useful intelligent assistants for work, it is important to better understand the work tasks that people are performing. Based on a survey of 401 respondents' daily tasks and activities in a work setting, we present a classification of work-related tasks, and analyze their key characteristics, including the frequency of their self-reported tasks, the environment in which they undertake the tasks, and which, if any, electronic devices are used. We also investigate the cyber, physical, and social aspects of tasks. Finally, we reflect on how intelligent assistants could influence and help people in a work environment to complete their tasks, and synthesize our findings to provide insight on the future of intelligent assistants in support of amplifying personal productivity.

Context-aware recommendation has attracted significant attentions over online sites due to its smart context adaption in improving recommendation quality. However, the user's instant contexts do not follow his/her regular user behaviour patterns, thus have not been well captured for advanced personalization of recommendation generation. In this work, we propose an Interactive Context-Aware Recommender System (I-CARS), which allows users to interact and present their needs, so the system can personalize and refine user preferences. I-CARS iteratively asks a question to a user to trigger feedback in term of her recent contexts and incorporates the response to recommend items most likely satisfying his/her instant interests. Specifically, we first propose a Personalized Weighted Context-Aware Matrix Factorization (PW-CAMF) that enables the personalization of important contexts for each user. Then we propose two question selection strategies that exploit user preferences through feedback. We have conducted comprehensive experiments over two real datasets. The experimental results prove the effectiveness of our I-CARS system compare to existing competitors.

We learn user preferences from ratings and reviews by using multi-task learning (MTL) of rating prediction and summarization of item reviews. Reviews of an item tend to describe detailed user preferences (e.g., the cast, genre, or screenplay of a movie). A summary of such a review or a rating describes an overall user experience of the item. Our objective is to learn latent vectors which are shared across rating prediction and review summary generation. Additionally, the learned latent vectors and the generated summary act as explanations for the recommendation. Our MTL-based approach J3R uses a multi-layer perceptron for rating prediction, combined with pointer-generator networks with attention mechanism for the summarization component. We provide empirical evidence for joint learning of rating prediction and summary generation being beneficial for recommendation by conducting experiments on the Yelp dataset and six domains of the Amazon 5-core dataset. Additionally, we provide two ways of explanations visualizing (a) the user vectors on different topics of a domain, computed from our J3R approach and (b) a ten-word review summary of a review and the attention highlights generated on the review based on the user-item vectors

Recommender systems aim to suggest relevant items to users among a large number of available items. They have been successfully applied in various industries, such as e-commerce, education and digital health. On the other hand, clustering approaches can help the recommender systems to group users into appropriate clusters, which are considered as neighborhoods in prediction process. Although it is a fact that preferences of users vary over time, traditional clustering approaches fail to consider this important factor. To address this problem, a social recommender system is proposed in this paper, which is based on a temporal clustering approach. Specifically, the temporal information of ratings provided by users on items and also social information among the users are considered in the proposed method. Experimental results on a benchmark dataset show that the quality of recommendations based on the proposed method is significantly higher than the state-of-the-art methods in terms of both accuracy and coverage metrics.

Context-aware recommendation has emerged as perhaps the most popular service over online sites, and has seen applications to domains as diverse as entertainment, e-business, e-health and government services. There has been recent significant progress on the quality and scalability of recommender systems. However, we believe that different target users concern different contexts when they select an online item, which can greatly affect the quality of recommendation, and have not been investigated yet. In this paper, we propose a new type of recommender system, Declarative Context-Aware Recommender System (D-CARS), which enables the personalization of the contexts exploited for each target user by automatically analysing the viewing history of users. First, we propose a novel User-Window Non-negative Matrix Factorization topic model (UW-NMF) that adaptively identifies the significant contexts of users and constructs user profiles in a personalized manner. Then, we design a novel declarative context-aware recommendation algorithm that exploits the user context preference to identify a group of item candidates and its context distribution, based on a Subspace Ensemble Tree Model (SETM), which is constructed in the identified context subspace for item recommendation. Finally, we propose an algorithm that incrementally maintains our SETM model. Extensive experiments are conducted to prove the high effectiveness and efficiency of our D-CARS system.

Traditionally, recommender systems modelled the physical and cyber contextual influence on people's moving, querying, and browsing behaviors in isolation. Yet, searching, querying, and moving behaviors are intricately linked, especially indoors. Here, we introduce a tripartite location-query-browse graph (LQB) for nuanced contextual recommendations. The LQB graph consists of three kinds of nodes: locations, queries, and Web domains. Directed connections only between heterogeneous nodes represent the contextual influences, while connections of homogeneous nodes are inferred from the contextual influences of the other nodes. This tripartite LQB graph is more reliable than any monopartite or bipartite graph in contextual location, query, and Web content recommendations. We validate this LQB graph in an indoor retail scenario with extensive dataset of three logs collected from over 120,000 anonymized, opt-in users over a 1-year period in a large inner-city mall in Sydney, Australia. We characterize the contextual influences that correspond to the arcs in the LQB graph, and evaluate the usefulness of the LQB graph for location, query, and Web content recommendations. The experimental results show that the LQB graph successfully captures the contextual influence and significantly outperforms the state of the art in these applications.

We analyze 18‐ million rows of Wi‐Fi access logs collected over a 1‐year period from over 120,000 anonymized users at an inner city shopping mall. The anonymized data set gathered from an opt‐in system provides users' approximate physical location as well as web browsing and some search history. Such data provide a unique opportunity to analyze the interaction between people's behavior in physical retail spaces and their web behavior, serving as a proxy to their information needs. We found that (a) there is a weekly periodicity in users' visits to the mall; (b) people tend to visit similar mall locations and web content during their repeated visits to the mall; (c) around 60% of registered Wi‐Fi users actively browse the web, and around 10% of them use Wi‐Fi for accessing web search engines; (d) people are likely to spend a relatively constant amount of time browsing the web while the duration of their visit may vary; (e) the physical spatial context has a small, but significant, influence on the web content that indoor users browse; and (f) accompanying users tend to access resources from the same web domains.

Currently, large amounts of Wi-Fi access logs are collected in diverse indoor environments, but cannot be widely used for fine-grained spatio-temporal analysis due to coarse positioning. We present a Log-based Differential (D-Log) scheme for post-hoc localization based on differentiated location estimates obtained from large-scale Access Point (AP) logs of WiFi connectivity traces, which can be used for data analysis and knowledge discovery of visitor behaviours. Specifically, the location estimates are calculated by utilizing a combination of Received Signal Strength Indicator (RSSI) records from two neighbouring APs. D-Log exploits real-world industry WiFi logs where RSSI data sampled at low rates from single AP sources are recorded in each connectivity trace. The approach is independent of device and network infrastructure type. D-Log is evaluated using WiFi logs collected from controlled environment as well as real-world uncontrolled public indoor spaces, which includes discrete single-AP RSSI traces of around 100,000 mobile devices over a one-year period. The experiment results indicate that, despite of the challenges with the infrequent sampling rate and the limitations of the data that only records RSSI from single AP sources in each instance, D-Log performs comparatively well to the state-of-the-art RSSI-based localization methods and presents a viable alternative for many application areas where high-accuracy positioning infrastructure may not be cost effective or where positioning applications are considered on legacy information infrastructure.

Summary Recommendations based on offline data processing has attracted increasing attention from both research communities and IT industries. The recommendation techniques could be used to explore huge volumes of data, identify the items that users probably like, translate the research results into real‐world applications and so on. This paper surveys the recent progress in the research of recommendations based on offline data processing, with emphasis on new techniques (such astemporal recommendation,graph‐based recommendationandtrust‐based recommendation), new features (such asserendipitous recommendation) and new research issues (such astag recommendationandgroup recommendation). We also provide an extensive review of evaluation measurements, benchmark data sets and available open source tools. Finally, we outline some existing challenges for future research. Copyright © 2014 John Wiley & Sons, Ltd.

Tagging recommender systems allow Internet users to annotate resources with personalized tags. The connection among users, resources and these annotations, often called a folksonomy, permits users the freedom to explore tags, and to obtain recommendations. Releasing these tagging datasets accelerat es both commercial and research work on recommender systems. However, tagging recommender systems has been confronted with serious privacy concerns because adversaries may re-identify a user and her/his sensitive information from the tagging dataset using a little background information. Recently, several private techniques have been proposed to address the problem, but most of them lack a strict privacy notion, and can hardly resist the number of possible attacks. This paper proposes an private releasing algorithm to perturb users’ profile in a strict privacy notion, differential privacy, with the goal of preserving a user’s identity in a tagging dataset. The algorithm includes three privacy-preserving operations: Private Tag Clustering is used to shrink the randomized domain and Private Tag Selection is then applied to find the most suitable replacement tags for the original tags. To hide the numbers of tags, the third operation, Weight Perturbation, finally adds Laplace noise to the weight of tags. We present extensive experimental results on two real world datasets, De.licio.us and Bibsonomy. While the personalization algorithm is successful in both cases, our results further suggest the private releasing algorithm can successfully retain the utility of the datasets while preserving users’ identity.

We report a preliminary study of mobile Web behaviour in a large indoor retail space. By analysing a Web log collected over a 1 year period at an inner city shopping mall in Sydney, Australia, we found that 1) around 60% of registered Wi-Fi users actively browse the Internet, and the rest 40% do not, with around 10% of these users using Web search engines. Around 70% of this Web activity in the investigated mall come from frequent visitors; 2) the content that indoor users search for is different from the content they consume while browsing; 3) the popularity of future indoor search queries can be predicted with a simple theoretical model based on past queries treated as a weighted directed graph. The work described in this paper underpins applications such as the prediction of users' information needs, retail recommendation systems, and improving the mobile Web search experience.

Wi-Fi and smartphone based location tracking technologies have been widely applied in location based services (LBS). This research is a part of an Australian Research Council (ARC) project required by a large global shopping mall company located in Australia. It aims to develop an effective customer tracking approach for acquiring shopping behavior of customers and providing them better services. Currently, the log data provided by the company only recorded one Wi-Fi connection at a time for one smartphone user. This makes most of the conventional tracking and positioning methods inapplicable. In order to solve this problem, a new approach was developed, which takes advantage of both the cell of origin (CoO) method and the fingerprinting method. Its implementation mainly includes two phases: 1) the cell determination phase, including partitioning of the floors of the shopping mall using a Voronoi diagram and determining a unique cell for each wireless access point (AP), adjusting the cells manually according to the real floor layout, obstructions as well as an on-site observing and calibrating process. An AP-shop/room matching table and other relevant information are stored in a database; and 2) the user location tracking phase, including determination of the cell that the customer locates in according to the detected AP linked with the customer's mobile phone. A back-tracing method is also used to improve the tracking performance. This new approach was tested in a large shopping-mall-like space covering an area of 1000 square meters, containing 8 simulated shops and 11 APs. The results showed a 96% correct detection rate whilst the 4% erroneous detections occurred when a user was near the cell boundaries. This new approach is also easy to implement and integrate with other location tracking systems.

Modelling the temporal dynamics of personal preferences is still under-developed despite the rapid development of personalization. In this paper, we observe that the user preference styles tend to change regularly following certain patterns in the co

In the context of collaborative filtering, the well-known data sparsity issue makes two like-minded users have little similarity, and consequently renders the k nearest neighbour rule inapplicable. In this paper, we address the data sparsity problem in the neighbourhood-based CF methods by proposing an Adaptive-Maximum imputation method (AdaM). The basic idea is to identify an imputation area that can maximize the imputation benefit for recommendation purposes, while minimizing the imputation error brought in. To achieve the maximum imputation benefit, the imputation area is determined from both the user and the item perspectives; to minimize the imputation error, there is at least one real rating preserved for each item in the identified imputation area. A theoretical analysis is provided to prove that the proposed imputation method outperforms the conventional neighbourhood-based CF methods through more accurate neighbour identification. Experiment results on benchmark datasets show that the proposed method significantly outperforms the other related state-of-the-art imputation-based methods in terms of accuracy.

As a popular technique in recommender systems, Collaborative Filtering (CF) has received extensive attention in recent years. However, its privacy-related issues, especially for neighborhood-based CF methods, can not be overlooked. The aim of this study is to address the privacy issues in the context of neighborhood-based CF methods by proposing a Private Neighbor Collaborative Filtering (PNCF) algorithm. The algorithm includes two privacy-preserving operations: Private Neighbor Selection and Recommendation-Aware Sensitivity. Private Neighbor Selection is constructed on the basis of the notion of differential privacy to privately choose neighbors. Recommendation-Aware Sensitivity is introduced to enhance the performance of recommendations. Theoretical and experimental analysis are provided to show the proposed algorithm can preserve differential privacy while retaining the accuracy of recommendations.

Tagging recommender systems allow Internet users to annotate resources with personalized tags. The connection among users, resources and these annotations, often called afolksonomy, permits users the freedom to explore tags, and to obtain recommendations. Releasing these tagging datasets accelerates both commercial and research work on recommender systems. However, adversaries may re-identify a user and her/his sensitivity information from the tagging dataset using a little background information. Recently, several private techniques have been proposed to address the problem, but most of them lack a strict privacy notion, and can hardly resist the number of possible attacks. This paper proposes an private releasing algorithm to perturb users' profile in a strict privacy notion, differential privacy, with the goal of preserving a user's identity in a tagging dataset. The algorithm includes three privacy preserving operations: Private Tag Clustering is used to shrink the randomized domain and Private Tag Selection is then applied to find the most suitable replacement tags for the original tags. To hide the numbers of tags, the third operation, Weight Perturbation, finally adds Lap lace noise to the weight of tags We present extensive experimental results on two real world datasets, Delicious and Bibsonomy. While the personalization algorithmis successful in both cases.

As one of the biggest challenges in research on recommender systems, the data sparsity issue is mainly caused by the fact that users tend to rate a small proportion of items from the huge number of available items. This issue becomes even more problematic for the neighborhood-based collaborative filtering (CF) methods, as there are even lower numbers of ratings available in the neighborhood of the query item. In this paper, we aim to address the data sparsity issue in the context of neighborhood-based CF. For a given query (user, item), a set of key ratings is first identified by taking the historical information of both the user and the item into account. Then, an auto-adaptive imputation (AutAI) method is proposed to impute the missing values in the set of key ratings. We present a theoretical analysis to show that the proposed imputation method effectively improves the performance of the conventional neighborhood-based CF methods. The experimental results show that our new method of CF with AutAI outperforms six existing recommendation methods in terms of accuracy.

Two rating patterns exist in the user × item rating matrix and influence each other: the personal rating patterns are hidden in each user's entire rating history, while the global rating patterns are hidden in the entire user × item rating matrix. In this paper, a Rating Pattern Subspace is proposed to model both of the rating patterns simultaneously by iteratively refining each other with an EM-like algorithm. Firstly, a low-rank subspace is built up to model the global rating patterns from the whole user × item rating matrix, then, the projection for each user on the subspace is refined individually based on his/her own entire rating history. After that, the refined user projections on the subspace are used to improve the modelling of the global rating patterns. Iteratively, we can obtain a well-trained low-rank Rating Pattern Subspace, which is capable of modelling both the personal and the global rating patterns. Based on this subspace, we propose a RapSVD algorithm to generate Top-N recommendations, and the experiment results show that the proposed method can significantly outperform the other state-of-the-art Top-N recommendation methods in terms of accuracy, especially on long tail item recommendations.

As each user tends to rate a small proportion of available items, the resulted Data Sparsity issue brings significant challenges to the research of recommender systems. This issue becomes even more severe for neighborhood-based collaborative filtering methods, as there are even lower numbers of ratings available in the neighborhood of the query item. In this paper, we aim to address the Data Sparsity issue in the context of the neighborhood-based collaborative filtering. Given the (user, item) query, a set of key ratings are identified, and an auto-adaptive imputation method is proposed to fill the missing values in the set of key ratings. The proposed method can be used with any similarity metrics, such as the Pearson Correlation Coefficient and Cosine-based similarity, and it is theoretically guaranteed to outperform the neighborhood-based collaborative filtering approaches. Results from experiments prove that the proposed method could significantly improve the accuracy of recommendations for neighborhood-based Collaborative Filtering algorithms.

In this paper, we observe that the user preference styles tend to change regularly following certain patterns. Therefore, we propose a Preference Pattern model to capture the user preference styles and their temporal dynamics, and apply this model to improve the accuracy of the Top-N recommendation. Precisely, a preference pattern is defined as a set of user preference styles sorted in a time order. The basic idea is to model user preference styles and their temporal dynamics by constructing a representative subspace with an Expectation-Maximization (EM)-like algorithm, which works in an iterative fashion by refining the global and the personal preference styles simultaneously. Then, the degree which the recommendations match the active user's preference styles, can be estimated by measuring its reconstruction error from its projection on the representative subspace. The experiment results indicate that the proposed model is robust to the data sparsity problem, and can significantly outperform the state-of-the-art algorithms on the Top-N recommendation in terms of accuracy.