Fairness in Recommender Systems

Michael D. Ekstrand, Anubrata Das, Robin Burke, Fernando Diaz

Abstract: Recent research has added the idea of fairness to the suite of concerns beyond accuracy or user satisfaction that recommender systems researchers and practitioners consider in their work. Recommender systems pose unique challenges for investigating the fairness and non-discrimination concepts that have been developed in other machine learning literature. The multistakeholder nature of recommender applications, the ranked outputs, the centrality of personalization, and the role of user response complicate the problem of identifying precisely what types and operationalizations of fairness may be relevant. In this chapter, we lay out various ways a recommender system may be unfair and provide a conceptual framework for identifying the fairness that arise in an application and designing a project to assess and mitigate them. We then survey the literature to date on fair recommendation and provide pointers to other research on algorithmic fairness we believe is a promising basis for improving the fairness of recommender systems.