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 .
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.
In the ever-evolving digital audio landscape, Spotify, well-known for its music and talk content, has recently introduced audiobooks to its vast user base. While promising, this move presents significant challenges for personalized recommendations. Unlike music and podcasts, audiobooks, initially available for a fee, cannot be easily skimmed before purchase, posing higher stakes for the relevance of recommendations. Furthermore, introducing a new content type into an existing platform confronts extreme data sparsity, as most users are unfamiliar with this new content type. Lastly, recommending content to millions of users requires the model to react fast and be scalable. To address these challenges, we leverage podcast and music user preferences and introduce 2T-HGNN, a scalable recommendation system comprising Heterogeneous Graph Neural Networks (HGNNs) and a Two Tower (2T) model. This novel approach uncovers nuanced item relationships while ensuring low latency and complexity. We decouple users from the HGNN graph and propose an innovative multi-link neighbor sampler. These choices, together with the 2T component, significantly reduce the complexity of the HGNN model. Empirical evaluations involving millions of users show significant improvement in the quality of personalized recommendations, resulting in a +46% increase in new audiobooks start rate and a +23% boost in streaming rates. Intriguingly, our model's impact extends beyond audiobooks, benefiting established products like podcasts.
Within the objectives of the H2020 DIAMOND project, the paper investigates women’s needs and expectations as users of the bike-sharing service managed by Syndicat Mixte Autolib et Velib Métropole in the territory of Paris Region-Petite Couronne (France). The paper presents a thematic literature review focused on gender inclusion in bike-sharing schemes. The proposed methodological approach is based on (i) Geographic Information Systems for the analysis of geolocated open datasets related to land, sociodemographic and mobility characteristics of the areas surrounding each docking stations. This was aimed at identifying a short list of suitable bike-sharing docking stations, which were further characterized through: (ii) structured proprietary data focused on travel demand; (iii) onsite observations focused on universal design indicators; (iv) survey questionnaires focused on women’s concerns, needs and expectations; and (v) social media data from Twitter focused on the opinion of the end-users. Results showed that women use the VELIB’s bike-sharing service much less than men (about 30% of the total number of users), since they are more concerned about the following issues: accessibility (e.g., availability of bikes at the docking stations, distance to the nearest station, type and quality of the cycle paths); safety and security (e.g., perception of danger and insecurity while cycling and using the current bicycle infrastructures); social constraints (e.g., perceptions and cultural stigmatization associated with cycling and bike-sharing); weather and topography (e.g., impact of weather and the urban terrain on cycling and bike-sharing). The final aim of the H2020 DIAMOND project is to support the definition of guidelines and policies for the inclusion of women’s needs in the design of future bike-sharing services.