Graph models that manipulate the frequency spectrum of user-item interactions to separate preference signals from noise often report significant improvements, but concerns about evaluation rigor and reproducibility persist. We conduct a reproducibility and replicability study that examines three major families: (i) spectral denoising methods, (ii) graph signal processing (GSP) models, and (iii) spectral propagation approaches. Reproducing published pipelines reveals a polarized landscape: while several works are fully reproducible, others rely on flawed metric implementations and incomplete hyperparameter disclosures. In particular, we observe systematic inflation of Recall in the spectral denoising methods due to an implementation error, and theoretically invalid ranking metrics in GSP models due to unordered prediction lists; conversely, the graph filtering models are consistently reproducible. Beyond reproduction, we establish a unified evaluation protocol on four datasets with consistent splits and hyperparameter optimization for all baselines, showing that strong classical methods (e.g., SLIM, Item-kNN) remain highly competitive and that no single spectral model dominates across domains. We further analyze robustness under varying data sparsity and assess beyond-accuracy properties, finding that spectral filtering often improves catalog exploration even when accuracy gains are marginal. Our code is available at https://github.com/sisinflab/Mind_the_Metric_SIGIR-26.

Price is a key determinant of user purchase behavior in e-commerce. As traditional recommendation techniques typically fail to account for price information, Price-Aware Recommender Systems have emerged in response, aiming to explicitly incorporate price as a core feature to enhance the overall quality and relevance of recommendations. Recent advances include session-based approaches such as CoHHN (Heterogeneous Hypergraphs), PASBR (Graph Neural Network), and the collaborative-filtering approach PUP (Graph Convolutional Networks). All reporting improved performance on their respective evaluation datasets. However, these methods have never been compared under controlled experimental conditions, limiting our understanding of their relative performance. This work addresses this gap through a reproducibility study over CoHHN, PASBR, and PUP. We successfully conducted the first cross-evaluation across six public datasets using standardized metrics (HR@20, MRR@20, NDCG@20). Our findings indicate that CoHHN provides the strongest and most consistent performance across evaluation metrics and datasets, while PASBR shows domain-limited competitiveness and PUP exhibits the weakest generalization, highlighting CoHHN's heterogeneous hypergraph architecture as the most robust framework for Price-Aware Recommendation. Source code available at https://anonymous.4open.science/r/reproducibility-sigir2026-E2CE.

Online ranking systems face the challenge of incorporating temporal information while maintaining ethical principles that avoid user profiling. Traditional approaches either ignore temporal dynamics entirely or rely on user reputation schemes that raise privacy and bias concerns. In this work, we propose T-UARS (Temporal User-Agnostic Ranking System), a methodological framework that integrates temporal awareness into ranking systems without assigning scores to users. Our approach uses exponential decay weighting and adaptive change detection to distinguish between anomalous ratings and genuine temporal patterns in item evaluations. We evaluate T-UARS on four datasets with varying temporal characteristics, showing that our framework maintains robust performance across domains, adapts to temporal signals in 10–44% of items, and remains resistant to manipulation while being computationally efficient. Our results show that temporal awareness and ethical ranking principles can be successfully combined, providing a foundation for privacy-preserving ranking systems that adapt to changing information landscapes. Source code: https://tinyurl.com/bp87nk8f.

Self-explaining recommenders enhance user trust by providing justifications for their suggestions. Sequence-aware models have advanced the field by leveraging user interaction history to personalize recommendations and explanations. However, generative models often struggle with sparse data, producing repetitive or irrelevant explanations. This paper explores the optimal methods for infusing rich textual information from past user interactions directly into the item embeddings to feed a user reasoning path leading to personalized explanations. We conduct a comprehensive analysis of various techniques, including: (1) multiple text aggregation strategies to pool fine-grained attributed item opinions into user-aggregated item text representations; (2) several fusion mechanisms to combine text and collaborative modalities, from early fusion to a late fusion approach within the Transformer architecture; and (3) different training regimes for explanation generation. Experiments on three real-world datasets demonstrate which steps to follow in order to successfully leverage textual information into a sequence-aware explainable recommendation model and boost recommendation performance as well as explanation quality.

Online item ranking systems are crucial for digital marketplaces, directly influencing user experience and vendor revenue. Traditional reputation-based ranking systems weight ratings according to user reputation scores. They have proven effective against manipulation but raise significant ethical concerns regarding user discrimination and privacy, and may raise concerns under emerging regulatory frameworks in certain application contexts. While a user-agnostic ranking system was recently introduced as an alternative approach that uses statistical filtering instead of user reputation scores, its theoretical foundations and resistance to bribing strategies remained unexplored. In this article, we provide the first comprehensive theoretical analysis of user-agnostic ranking system’s robustness properties. We establish formal bounds on bribing resistance by proving three key properties: strategy composition conditions, profitability constraints, and statistical validity requirements. Our theoretical framework demonstrates that profitable bribing strategies in this class of system must satisfy strict statistical conditions, making manipulation more difficult than in reputation-based systems. Experimental evaluation on three real-world datasets confirms our theoretical findings, showing that user-agnostic ranking systems can achieve superior bribing resistance while maintaining comparable effectiveness and efficiency.

In the rapidly evolving landscape of technology and sustainability, leveraging Recommender Systems has emerged as a powerful tool for driving positive change. With a foundation in AI and data analytics, Recommender Systems can be effective in various domains, from e-commerce to energy management, inclusion and well-being. By harnessing the power of recommendation algorithms under a multi-stakeholder perspective, organizations and researchers can guide users towards more sustainable choices and behaviors, contributing to broader environmental and social goals. With this aim, our workshop provides a unique platform for researchers, practitioners, and platform owners to explore the integration of sustainability principles into Recommender Systems. Through presentations, discussions, and panels, participants can explore the theoretical foundations, practical implementations, and ethical and environmental considerations of sustainable Recommender Systems. By fostering collaboration and knowledge exchange, the workshop aims to catalyze innovation and inspire collective action towards a more sustainable future.

Multi-objective recommender systems (MORS) aim to optimize multiple criteria while generating recommendations, such as relevance, novelty, diversity, or exploration.These algorithms are based on the assumption that an operationalization of these criteria (i.e., translating abstract goals into measurable metrics), will reflect how users perceive them.Nevertheless, such beliefs are rarely rigorously evaluated, which can lead to a mismatch between algorithmic goals and user satisfaction.Moreover, if users are allowed to control the RS via their propensities towards such objectives, the misconceptions may further impact users' trust and engagement.To characterize this problem, we conduct a large user study focusing on recommender systems in two domains: books and movies.Part of the study is focused on how users perceive different recommendation objectives, which we compared with well-established metrics aiming at the same objectives.We found that despite such metrics correlating to some extent with users' perceptions, the mapping is far from perfect.Moreover, we also report on conceptual-level differences in users' understanding of RS objectives and how this affects the results.Study data are available from https://osf.io/2n9mf/.

The governance of recommender systems (RSs) in very large online platforms (VLOPs) is expected to change significantly under the Digital Services Act (DSA), which imposes new obligations on transparency and user control.However, beyond legal compliance, a critical question remains: How can recommender systems be redesigned to genuinely empower users and foster meaningful personalization?This paper addresses this question by analyzing how three major short-video platforms-Instagram, TikTok, and YouTube-have implemented the DSA requirements for RSs.By reviewing their audit reports, systemic risk assessments, and compliance strategies, we evaluate the extent to which current approaches enhance user autonomy and control over content exposure.Building on this analysis, we outline a perspective for the future of VLOPs' RSs grounded in speculative design.We argue that meaningful personalization should integrate algorithmic choice, balancing proportionality and granularity in RS customization, and content curation, ensuring diversity and authoritativeness to mitigate systemic risks.By bridging legal analysis, platform governance, and user-centered design, this paper outlines actionable pathways for aligning technical developments with regulatory objectives.Our findings contribute to interdisciplinary research on RSs by highlighting how platforms can move beyond minimal compliance toward a model that prioritizes user empowerment and content pluralism.

Group accommodation booking forces travelers to coordinate externally through messaging apps and informal voting, missing opportunities for transparent preference alignment. We present PRISM, an interactive group recommender system that transforms opaque recommendation processes into transparent collaborative visual experiences. PRISM employs a two-phase interaction paradigm: individual preference elicitation through conversational AI, followed by collaborative decision-making via bivariate map preference visualization. A controlled user study with 6 pairs shows PRISM enhances transparency (+1.83 on 5-point scale), consensus building (+2.0), and reduces conformity pressure compared to traditional approaches and interfaces.

Music streaming platforms reduce rich listening experiences to algorithmic black boxes, overlooking personal narratives that make music meaningful. We present Blooming Beats, an explainable recommender system that transforms Spotify listening data into visual narratives using Data Humanism principles. The system embodies TRACE principles: Transparency through visual explanations, Context-awareness by integrating personal context, and Empathy by matching listening stories rather than user profiles. A user study with 8 participants exploring a decade of listening data shows that narrative-driven visualization suggests potential for enhancing transparency and engagement.

Designing search and recommendation models that are both efficient and effective has long been a central objective for both industry professionals and academic researchers. Yet, growing evidence highlights how models trained on historical data can reinforce pre-existing biases, potentially leading to harmful outcomes. Addressing these challenges by defining, evaluating, and mitigating bias across development workflows is a crucial step toward the responsible deployment of search and recommendation models in practice. The BIAS 2025 workshop seeks to gather innovative research and foster a shared space for dialogue among researchers and practitioners committed to advancing this fundamental direction. Workshop website: https://biasinrecsys.github.io/sigir2025/.

Autism Spectrum Disorder (ASD) affects sensory perception, making spatial exploration difficult. Recommender systems can assist ASD users by suggesting Points of Interest (POIs) aligned with their sensory preferences. However, demographic constraints, difficulties in engaging ASD users, and the complexity of obtaining sensory data position POI recommendation for ASD people as a low-resource problem. In this paper, we identify key challenges in developing such systems and present our ongoing efforts. Using a local ASD center as a use case, we are developing a structured user involvement protocol. From the limited data, we are deriving knowledge graphs (KGs) to model preferences and sensory aspects. We are then exploring KG-based techniques to generate paths from users to POIs to suggest. With psychologists, we are refining the paths structure to match varying complexity levels and translate them into natural language accessible for people with ASD.

Protecting user privacy in recommender systems is crucial for fostering trust in marketplaces. In this paper, we propose a privacy-preserving framework that integrates public seller rankings into personalized recommendations without exposing sensitive user preferences. By utilizing ''seller representative users'' (encoding seller item rankings) and a novel recommendation mechanism, the framework preserves privacy while ensuring robust ranking accuracy. Our approach is validated on multiple use cases extracted from real-world datasets, showing its effectiveness across varying marketplace configurations. This framework is suited for real-world applications, such as e-commerce platforms, where it can enhance user trust, protect sensitive data, and improve engagement by transparently balancing personalization and privacy.

Nowadays, research into personalization has been focusing on explainability and fairness. Several approaches proposed in recent works are able to explain individual recommendations in a post-hoc manner or by explanation paths. However, explainability techniques applied to unfairness in recommendation have been limited to finding user/item features mostly related to biased recommendations. In this article, we devised a novel algorithm that leverages counterfactuality methods to discover user unfairness explanations in the form of user-item interactions. In our counterfactual framework, interactions are represented as edges in a bipartite graph, with users and items as nodes. Our bipartite graph explainer perturbs the topological structure to find an altered version that minimizes the disparity in utility between the protected and unprotected demographic groups. Experiments on four real-world graphs coming from various domains showed that our method can systematically explain user unfairness on three state-of-the-art GNN-based recommendation models. Moreover, an empirical evaluation of the perturbed network uncovered relevant patterns that justify the nature of the unfairness discovered by the generated explanations. The source code and the preprocessed data sets are available at https://github.com/jackmedda/RS-BGExplainer .

As education adopts digital platforms, the vast amount of information from various sources, such as learning management systems and learning object repositories, presents challenges in navigation and elaboration. Traditional interfaces involve a steep learning curve, limited user accessibility, and lack flexibility. Language models alone cannot address these issues as they do not have access to structured information specific to the educational organization. In this paper, we propose EDGE (EDucational knowledge Graph Explorer), a natural language interface that uses knowledge graphs to organize educational information. EDGE translates natural language requests into queries and converts the results back into natural language responses. We show EDGE's versatility using knowledge graphs built from public datasets, providing example interactions of different stakeholders. Demo video: https://u.garr.it/eYq63.

One way to increase trust among users towards recommender systems is to provide the recommendation along with a textual explanation. In the literature, extraction-based, generation-based, and, more recently, hybrid solutions based on retrieval-augmented generation have been proposed to tackle the problem of text-based explainable recommendation. However, the use of different datasets, preprocessing steps, target explanations, baselines, and evaluation metrics complicates the reproducibility and state-of-the-art assessment of previous work among different model categories for successful advancements in the field. Our aim is to provide a comprehensive analysis of text-based explainable recommender systems by setting up a well-defined benchmark that accommodates generation-based, extraction-based, and hybrid approaches. Also, we enrich the existing evaluation of explainability and text quality of the explanations with a novel definition of feature hallucination. Our experiments on three real-world datasets unveil hidden behaviors and confirm several claims about model patterns. Our source code and preprocessed datasets are available at https://github.com/alarca94/text-exp-recsys24.

Current recommendation methods based on knowledge graphs rely on entity and relation representations for several steps along the pipeline, with knowledge completion and path reasoning being the most influential. Despite their similarities, the most effective representation methods for these steps differ, leading to inefficiencies, limited representativeness, and reduced interpretability. In this paper, we introduce KGGLM, a decoder-only Transformer model designed for generalizable knowledge representation learning to support recommendation. The model is trained on generic paths sampled from the knowledge graph to capture foundational patterns, and then fine-tuned on paths specific of the downstream step (knowledge completion and path reasoning in our case). Experiments on ML1M and LFM1M show that KGGLM beats twenty-two baselines in effectiveness under both knowledge completion and recommendation. Source code and pre-processed data sets are available at https://github.com/mirkomarras/kgglm.

Recent developments in recommendation have harnessed the collaborative power of graph neural networks (GNNs) in learning users’ preferences from user-item networks. Despite emerging regulations addressing fairness of automated systems, unfairness issues in graph collaborative filtering remain underexplored, especially from the consumer’s perspective. Despite numerous contributions on consumer unfairness, only a few of these works have delved into GNNs. A notable gap exists in the formalization of the latest mitigation algorithms, as well as in their effectiveness and reliability on cutting-edge models. This paper serves as a solid response to recent research highlighting unfairness issues in graph collaborative filtering by reproducing one of the latest mitigation methods. The reproduced technique adjusts the system fairness level by learning a fair graph augmentation. Under an experimental setup based on 11 GNNs, 5 non-GNN models, and 5 real-world networks across diverse domains, our investigation reveals that fair graph augmentation is consistently effective on high-utility models and large datasets. Experiments on the transferability of the fair augmented graph open new issues for future recommendation studies. Source code: https://github.com/jackmedda/FA4GCF.

In the rapidly evolving landscape of technology and sustainability, leveraging Recommender Systems has emerged as a powerful tool for driving positive change. With a foundation in AI and data analytics, Recommender Systems can be effective in various domains, from e-commerce to energy management and well-being. By harnessing the power of recommendation algorithms under a holistic perspective, organizations and researchers can guide users towards more sustainable choices and behaviors, contributing to broader environmental and social goals. With this aim, our workshop provides a unique opportunity for researchers, practitioners, and stakeholders to explore the integration of sustainability principles into Recommender Systems. Through presentations, discussions, and panels, participants explore the theoretical foundations, practical implementations, and ethical and environmental issues of sustainable Recommender Systems. By fostering collaboration and knowledge exchange, the workshop aims to catalyze innovation and inspire collective action towards a more sustainable future.

Nowadays a recommendation model should exploit additional information from both the user and item perspectives, in addition to utilizing user-item interaction data. Datasets are central in offering the required information for evaluating new models or algorithms. Although there are many datasets in the literature with user and item properties, there are several issues not covered yet: (i) it is difficult to perform cross-analysis of properties at user and item level as they are not related in most cases; and (ii) on top of that, in many occasions datasets do not allow analysis at different granularity levels. In this paper, we propose a new dataset in the music domain, named AMBAR, that tackles the above-mentioned issues. Besides detailing in depth the structure of the new dataset, we also show its application in contexts (i.e., multi-objective, fair, and calibrated recommendations) where both the effectiveness and the beyond-accuracy perspectives of recommendation are assessed.

Creating efficient and effective search and recommendation algorithms has been the main objective of industry practitioners and academic researchers over the years. However, recent research has shown how these algorithms trained on historical data lead to models that might exacerbate existing biases and generate potentially negative outcomes. Defining, assessing, and mitigating these biases throughout experimental pipelines is a primary step for devising search and recommendation algorithms that can be responsibly deployed in real-world applications. This workshop aims to collect novel contributions in this field and offer a common ground for interested researchers and practitioners. More information about the workshop is available at https://biasinrecsys.github.io/sigir2024/

Recommender systems (RS) rely on interaction data between users and items to generate effective results. Historically, RS aimed to deliver the most consistent (i.e., accurate) items to the trained user profiles. However, the attention towards additional (beyond-accuracy) quality criteria has increased tremendously in recent years. Both the research and applied models are being optimized for diversity, novelty, or fairness, to name a few. Naturally, the proper functioning of such optimization methods depends on the knowledge of users' propensities towards interacting with recommendations having certain quality criteria. However, so far, no dataset that captures such propensities exists. To bridge this research gap, we present SM-RS (single-objective + multi-objective recommendations dataset) that links users' self-declared propensity toward relevance, novelty, and diversity criteria with impressions and corresponding item selections. After presenting the dataset's collection procedure and basic statistics, we propose three tasks that are rarely available to conduct using existing RS datasets: impressions-aware click prediction, users' propensity scores prediction, and construction of recommendations proportional to the users' propensity scores. For each task, we also provide detailed evaluation procedures and competitive baselines. The dataset is available at https://osf.io/hkzje/.

Recommender systems (RecSys) solve personalisation problems and therefore heavily rely on personal data - demographics, user preferences, user interactions - each baring important privacy risks. It is also widely accepted that in RecSys performance and privacy are at odds, with the increase of one resulting in the decrease of the other. Among the diverse approaches in privacy enhancing technologies (PET) for RecSys, perturbation stands out for its simplicity and computational efficiency. It involves adding noise to sensitive data, thus hiding its real value from an untrusted actor. We reproduce and test a set of four randomization-based perturbation techniques developed by Batmaz and Polat \citebatmaz2016randomization for privacy preserving collaborative filtering. While the framework presents great advantages - low computational requirements, several useful privacy-enhancing parameters - the supporting paper lacks conclusions drawn from empirical evaluation. We address this shortcoming by proposing - in absence of an implementation by the authors - our own implementation of the obfuscation framework. We then develop an evaluation framework to test the main assumption of the reference paper - that RecSys privacy and performance are competing goals. We extend this study to understand how much we can enhance privacy, within reasonable losses of the RecSys performance. We reproduce and test the framework for the more realistic scenario where only implicit feedback is available, using two well-known datasets (MovieLens-1M and Last.fm-1K), and several state-of-the-art recommendation algorithms (NCF and LightGCN from the Microsoft Recommenders public repository).

Ranking systems are instrumental in shaping user experiences by determining the relevance and order of presented items. However, current approaches, particularly those revolving around user-centric reputation scoring, raise ethical concerns associated with scoring individuals. To counter such issues, in this paper, we introduce a novel item ranking system approach that strategically transitions its emphasis from scoring users to calculating item rankings relying exclusively on items' ratings information, to achieve the same objective. Experiments on three datasets show that our approach achieves higher effectiveness and efficiency than state-of-the-art baselines. Furthermore, the resulting rankings are more robust to spam and resistant to bribery, contributing to a novel and ethically sound direction for item ranking systems.

In recommendation literature, explainability and fairness are becoming two prominent perspectives to consider. However, prior works have mostly addressed them separately, for instance by explaining to consumers why a certain item was recommended or mitigating disparate impacts in recommendation utility. None of them has leveraged explainability techniques to inform unfairness mitigation. In this paper, we propose an approach that relies on counterfactual explanations to augment the set of user-item interactions, such that using them while inferring recommendations leads to fairer outcomes. Modeling user-item interactions as a bipartite graph, our approach augments the latter by identifying new user-item edges that not only can explain the original unfairness by design, but can also mitigate it. Experiments on two public data sets show that our approach effectively leads to a better trade-off between fairness and recommendation utility compared with state-of-the-art mitigation procedures. We further analyze the characteristics of added edges to highlight key unfairness patterns. Source code available at https://github.com/jackmedda/RS-BGExplainer/tree/cikm2023.

The proposed tutorial aims to familiarise the CIKM community with modern user profiling techniques that utilise Graph Neural Networks (GNNs). Initially, we will delve into the foundational principles of user profiling and GNNs, accompanied by an overview of relevant literature. We will subsequently systematically examine cutting-edge GNN architectures specifically developed for user profiling, highlighting the typical data utilised in this context. Furthermore, ethical considerations and beyond-accuracy perspectives, e.g. fairness and explainability, will be discussed regarding the potential applications of GNNs in user profiling. During the hands-on session, participants will gain practical insights into constructing and training recent GNN models for user profiling using open-source tools and publicly available datasets. The audience will actively explore the impact of these models through case studies focused on bias analysis and explanations of user profiles. To conclude the tutorial, we will analyse existing and emerging challenges in the field and discuss future research directions.

Self-explaining models are becoming an important perk of recommender systems, as they help users understand the reason behind certain recommendations, which encourages them to interact more often with the platform. In order to personalize recommendations, modern approaches make the model aware of the user behavior history for interest evolution representation. However, existing explainable recommender systems do not consider the past user history to further personalize the explanation based on the user interest fluctuation. In this work, we propose a SEQuence-Aware Explainable Recommendation model (SEQUER) that is able to leverage the sequence of user-item review interactions to generate better explanations while maintaining recommendation performance. Experiments validate the effectiveness of our proposal on multiple recommendation scenarios. Our source code and preprocessed datasets are available at https://github.com/alarca94/sequer-recsys23.

Multi-objective recommender systems (MORS) provide suggestions to users according to multiple (and possibly conflicting) goals. When a system optimizes its results at the individual-user level, it tailors them on a user’s propensity towards the different objectives. Hence, the capability to understand users’ fine-grained needs towards each goal is crucial. In this paper, we present the results of a user study in which we monitored the way users interacted with recommended items, as well as their self-proclaimed propensities towards relevance, novelty, and diversity objectives. The study was divided into several sessions, where users evaluated recommendation lists originating from a relevance-only single-objective baseline as well as MORS. We show that, despite MORS-based recommendations attracting fewer selections, their presence in the early sessions are crucial for users’ satisfaction in the later stages. Surprisingly, the self-proclaimed willingness of users to interact with novel and diverse items is not always reflected in the recommendations they accept. Post-study questionnaires provide insights on how to deal with this matter, suggesting that MORS-based results should be accompanied by elements that allow users to understand the recommendations, so as to facilitate the choice of whether a recommendation should be accepted or not. Detailed study results are available at https://bit.ly/looks-can-be-deceiving-repo.

Providing effective suggestions is of predominant importance for successful Recommender Systems (RSs). Nonetheless, the need of accounting for additional multiple objectives has become prominent, from both the final users’ and the item providers’ points of view. This need has led to a new class of RSs, called Multi-Objective Recommender Systems (MORSs). These systems are designed to provide suggestions by considering multiple (conflicting) objectives simultaneously, such as diverse, novel, and fairness-aware recommendations. In this work, we reproduce a state-of-the-art study on MORSs that exploits a reinforcement learning agent to satisfy three objectives, i.e., accuracy, diversity, and novelty of recommendations. The selected study is one of the few MORSs where the source code and datasets are released to ensure the reproducibility of the proposed approach. Interestingly, we find that some challenges arise when replicating the results of the original work, due to the nature of multiple-objective problems. We also extend the evaluation of the approach to analyze the impact of improving user-centered objectives of recommendations (i.e., diversity and novelty) in terms of algorithmic bias. To this end, we take into consideration both popularity and category of the items. We discover some interesting trends in the recommendation performance according to different evaluation metrics. In addition, we see that the multi-objective reinforcement learning approach is responsible for increasing the bias disparity in the output of the recommendation algorithm for those items belonging to positively/negatively biased categories. We publicly release datasets and codes in the following GitHub repository: https://github.com/sisinflab/MORS_reproducibility.

Modern user profiling approaches capture different forms of interactions with the data, from user-item to user-user relationships. Graph Neural Networks (GNNs) have become a natural way to model these behaviours and build efficient and effective user profiles. However, each GNN-based user profiling approach has its own way of processing information, thus creating heterogeneity that does not favour the benchmarking of these techniques. To overcome this issue, we present FairUP, a framework that standardises the input needed to run three state-of-the-art GNN-based models for user profiling tasks. Moreover, given the importance that algorithmic fairness is getting in the evaluation of machine learning systems, FairUP includes two additional components to (1) analyse pre-processing and post-processing fairness and (2) mitigate the potential presence of unfairness in the original datasets through three pre-processing debiasing techniques. The framework, while extensible in multiple directions, in its first version, allows the user to conduct experiments on four real-world datasets. The source code is available at https://link.erasmopurif.com/FairUP-source-code, and the web application is available at https://link.erasmopurif.com/FairUP.

Going beyond accuracy in the evaluation of a recommender system is an aspect that is receiving more and more attention. Among the many perspectives that can be considered, the impact of presentation bias is of central importance. Under presentation bias, the attention of the users to the items in a recommendation list changes, thus affecting their possibility to be considered and the effectiveness of a model. Page-wise within-subject studies are widely employed in the recommender systems literature to compare algorithms by displaying their results in parallel. However, no study has ever been performed to assess the impact of presentation bias in this context. In this paper, we characterize how presentation bias affects different layout options, which present the results in column- or row-wise fashion. Concretely, we present a user study where six layout variants are proposed to the users in a page-wise within-subject setting, so as to evaluate their perception of the displayed recommendations. Results show that presentation bias impacts users clicking behavior (low-level feedback), but not so much the perceived performance of a recommender system (high-level feedback). Source codes and raw results are available at https://tinyurl.com/PresBiasSIGIR2023.

Abstract Location-Based Social Networks stimulated the rise of services such as Location-based Recommender Systems. These systems suggest to users points of interest (or venues) to visit when they arrive in a specific city or region. These recommendations impact various stakeholders in society, like the users who receive the recommendations and venue owners. Hence, if a recommender generates biased or polarized results, this affects in tangible ways both the experience of the users and the providers’ activities. In this paper, we focus on four forms of polarization, namely venue popularity, category popularity, venue exposure, and geographical distance. We characterize them on different families of recommendation algorithms when using a realistic (temporal-aware) offline evaluation methodology while assessing their existence. Besides, we propose two automatic approaches to mitigate those biases. Experimental results on real-world data show that these approaches are able to jointly improve the recommendation effectiveness, while alleviating these multiple polarizations.

Recent approaches to behavioural user profiling employ Graph Neural Networks (GNNs) to turn users' interactions with a platform into actionable knowledge. The effectiveness of an approach is usually assessed with accuracy-based perspectives, where the capability to predict user features (such as gender or age) is evaluated. In this work, we perform a beyond-accuracy analysis of the state-of-the-art approaches to assess the presence of disparate impact and disparate mistreatment, meaning that users characterised by a given sensitive feature are unintentionally, but systematically, classified worse than their counterparts. Our analysis on two real-world datasets shows that different user profiling paradigms can impact fairness results. The source code and the preprocessed datasets are available at: https://github.com/erasmopurif/do_gnns_build_fair_models.

The goal of this tutorial is to present the RecSys community with recent advances on explainable recommender systems with knowledge graphs. We will first introduce conceptual foundations, by surveying the state of the art and describing real-world examples of how knowledge graphs are being integrated into the recommendation pipeline, also for the purpose of providing explanations. This tutorial will continue with a systematic presentation of algorithmic solutions to model, integrate, train, and assess a recommender system with knowledge graphs, with particular attention to the explainability perspective. A practical part will then provide attendees with concrete implementations of recommender systems with knowledge graphs, leveraging open-source tools and public datasets; in this part, tutorial participants will be engaged in the design of explanations accompanying the recommendations and in articulating their impact. We conclude the tutorial by analyzing emerging open issues and future directions. Website: https://explainablerecsys.github.io/recsys2022/.

Existing explainable recommender systems have mainly modeled relationships between recommended and already experienced products, and shaped explanation types accordingly (e.g., movie "x" starred by actress "y" recommended to a user because that user watched other movies with "y" as an actress). However, none of these systems has investigated the extent to which properties of a single explanation (e.g., the recency of interaction with that actress) and of a group of explanations for a recommended list (e.g., the diversity of the explanation types) can influence the perceived explaination quality. In this paper, we conceptualized three novel properties that model the quality of the explanations (linking interaction recency, shared entity popularity, and explanation type diversity) and proposed re-ranking approaches able to optimize for these properties. Experiments on two public data sets showed that our approaches can increase explanation quality according to the proposed properties, fairly across demographic groups, while preserving recommendation utility. The source code and data are available at https://github.com/giacoballoccu/explanation-quality-recsys.

Engaging all content providers, including newcomers or minority demographic groups, is crucial for online platforms to keep growing and working. Hence, while building recommendation services, the interests of those providers should be valued. In this paper, we consider providers as grouped based on a common characteristic in settings in which certain provider groups have low representation of items in the catalog and, thus, in the user interactions. Then, we envision a scenario wherein platform owners seek to control the degree of exposure to such groups in the recommendation process. To support this scenario, we rely on disparate exposure measures that characterize the gap between the share of recommendations given to groups and the target level of exposure pursued by the platform owners. We then propose a re-ranking procedure that ensures desired levels of exposure are met. Experiments show that, while supporting certain groups of providers by rendering them with the target exposure, beyond-accuracy objectives experience significant gains with negligible impact in recommendation utility.

The spread of online reviews and opinions and its growing influence on people’s behavior and decisions boosted the interest to extract meaningful information from this data deluge. Hence, crowdsourced ratings of products and services gained a critical role in business and governments. Current state-of-the-art solutions rank the items with an average of the ratings expressed for an item, with a consequent lack of personalization for the users, and the exposure to attacks and spamming/spurious users. Using these ratings to group users with similar preferences might be useful to present users with items that reflect their preferences and overcome those vulnerabilities. In this article, we propose a new reputation-based ranking system, utilizing multipartite rating subnetworks, which clusters users by their similarities using three measures, two of them based on Kolmogorov complexity. We also study its resistance to bribery and how to design optimal bribing strategies. Our system is novel in that it reflects the diversity of preferences by (possibly) assigning distinct rankings to the same item, for different groups of users. We prove the convergence and efficiency of the system. By testing it on synthetic and real data, we see that it copes better with spamming/spurious users, being more robust to attacks than state-of-the-art approaches. Also, by clustering users, the effect of bribery in the proposed multipartite ranking system is dimmed, comparing to the bipartite case.

Prediction bias is a well-known problem in classification algorithms, which tend to be skewed towards more represented classes. This phenomenon is even more remarkable in multi-label scenarios, where the number of underrepresented classes is usually larger. In light of this, we hereby present the Prediction Bias Coefficient (PBC), a novel measure that aims to assess the bias induced by label imbalance in multi-label classification. The approach leverages Spearman's rank correlation coefficient between the label frequencies and the F-scores obtained for each label individually. After describing the theoretical properties of the proposed indicator, we illustrate its behaviour on a classification task performed with state-of-the-art methods on two real-world datasets, and we compare it experimentally with other metrics described in the literature.

The impact of ranking systems on humans is an aspect that is getting a lot of attention. In this paper, we consider a class of algorithms, known as reputation-based ranking systems, which rank the items based on a reputation score automatically computed for each user. Recent literature introduced the concept of reputation independence, which considers a sensitive attribute of the users (such as gender or age) and makes the reputation scores independent from that attribute. Here, we show that if we consider a different sensitive attribute w.r.t. a user to introduce independence, reputation scores are still biased. To overcome this issue, we propose an approach to attain equity in the reputation scores computation, independently of any sensitive attribute that characterizes the users.

This tutorial presents recent advances on the assessment and mitigation of data and algorithmic bias in personalized rankings. We first introduce fundamental concepts and definitions associated with bias issues, covering the state of the art and describing real-world examples of how bias can impact ranking algorithms from several perspectives (e.g., ethics and system's objectives). Then, we continue with a systematic presentation of techniques to uncover, assess, and mitigate biases along the personalized ranking design process, with a focus on the role of data engineering in each step of the pipeline. Hands-on parts provide attendees with concrete implementations of bias mitigation algorithms, in addition to processes and guidelines on how data is organized and manipulated by these algorithms. The tutorial leverages open-source tools and public datasets, engaging attendees in designing bias countermeasures and in articulating impacts on stakeholders. We finally showcase open issues and future directions in this vibrant and rapidly evolving research area (Website: https://biasinrecsys.github.io/icde2021/).

Educational recommender systems channel most of the research efforts on the effectiveness of the recommended items. While teachers have a central role in online platforms, the impact of recommender systems for teachers in terms of the exposure such systems give to the courses is an under-explored area. In this paper, we consider data coming from a real-world platform and analyze the distribution of the recommendations w.r.t. the geographical provenience of the teachers. We observe that data is highly imbalanced towards the United States, in terms of offered courses and of interactions. These imbalances are exacerbated by recommender systems, which overexpose the country w.r.t. its representation in the data, thus generating unfairness for teachers outside that country. To introduce equity, we propose an approach that regulates the share of recommendations given to the items produced in a country (visibility) and the position of the items in the recommended list (exposure).

The goal of this tutorial is to provide the WSDM community with recent advances on the assessment and mitigation of data and algorithmic bias in recommender systems. We first introduce conceptual foundations, by presenting the state of the art and describing real-world examples of how bias can impact on recommendation algorithms from several perspectives (e.g., ethical and system objectives). The tutorial continues with a systematic showcase of algorithmic countermeasures to uncover, assess, and reduce bias along the recommendation design process. A practical part then provides attendees with implementations of pre-, in-, and post-processing bias mitigation algorithms, leveraging open-source tools and public datasets; in this part, tutorial participants are engaged in the design of bias countermeasures and in articulating impacts on stakeholders. We conclude the tutorial by analyzing emerging open issues and future directions in this rapidly evolving research area (Website: https://biasinrecsys.github.io/wsdm2021).

Recent observational studies highlight the importance of considering the interactions between users in the group recommendation process, but to date their integration has been marginal. In this article, we propose a collaborative model based on the social interactions that take place in a web-based conversational group recommender system. The collaborative model allows the group recommender to implicitly infer the different roles within the group, namely, collaborative and leader user(s). Moreover, it serves as the basis of several novel collaboration-based consensus strategies that integrate both individual and social interactions in the group recommendation process. A live-user evaluation confirms that our approach accurately identifies the collaborative and leader users in a group and produces more effective recommendations.

Recent literature on ranking systems (RS) has considered users' exposure when they are the object of the ranking. Although items are the object of reputation-based RS, users have a central role also in this class of algorithms. Indeed, when ranking the items, user preferences are weighted by how relevant this user is in the platform (i.e., their reputation). In this paper, we formulate the concept of disparate reputation (DR) and study if users characterized by sensitive attributes systematically get a lower reputation, leading to a final ranking that reflects less their preferences. We consider two demographic attributes, i.e., gender and age, and show that DR systematically occurs. Then, we propose mitigation, which ensures that reputation is independent of the users' sensitive attributes. Experiments on real-world data show that our approach can overcome DR and also improve ranking effectiveness.

Group recommender systems provide suggestions in contexts in which people operate in groups. The goal of this tutorial is to provide the RecSys audience with an overview on group recommendation. We will first formally introduce the problem of producing recommendations to groups, then present a survey based on the tasks performed by these systems. We will also analyze challenging topics like their evaluation, and present emerging aspects and techniques in this area. The tutorial will end with a summary that highlights open issues and research challenges.

The growth of the Web 2.0 has brought to a widespread use of social media systems. In particular, social bookmarking systems are a form of social media system that allows to tag bookmarks of interest for a user and to share them. The increasing popularity of these systems leads to an increasing number of active users and this implies that each user interacts with too many users ("social interaction overload"). In order to overcome this problem, we present a friend recommender system in the social bookmarking domain. Recommendations are produced by mining user behavior in a tagging system, analyzing the bookmarks tagged by a user and the frequency of each used tag. Experimental results highlight that, by analyzing both the tagging and bookmarking behavior of a user, our approach is able to mine preferences in a more accurate way, with respect to state-of-the-art approaches that consider only tags.