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. While fair recommendation shares many commonalities with the fairness constructs developed in other machine learning settings such as supervised classification, recommender systems pose unique challenges for investigating and applying the fairness and non-discrimination concepts that have been developed to date. The multistakeholder nature of recommender applications,the rank-based problem setting, 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, let alone measuring or ensuring them. In this chapter, we survey the extant literature on fair recommendation, as well as work from the rapidly-growing literature on algorithmic fairness that we believe will be a promising basis for future work on fair recommendation. We integrate this survey into a taxonomy of the various dimensions of the problem of fair recommendation and discuss several open problems and the steps that seem needed to address them.

Links

• The FnT paper contains and expanded treatment that broadens to information access generally
• The tutorial series