Fair Clustering for Diverse and Experienced Groups

The ability for machine learning to exacerbate bias has led to many algorithms centered on fairness. For example, fair clustering algorithms typically focus on balanced representation of protected attributes within clusters. Here, we develop a fair clustering variant where the input data is a hypergraph with multiple edge types, representing information about past experiences of groups of individuals. Our method is based on diversity of experience, instead of protected attributes, with a goal of forming groups that have both experience and diversity with respect to participation in edge types. We model this goal with a regularized edge-based clustering objective, design an efficient 2-approximation algorithm for optimizing the NP-hard objective, and provide bounds on hyperparameters to avoid trivial solutions. We demonstrate a potential application of this framework in online review platforms, where the goal is to curate sets of user reviews for a product type. In this context, "experience" corresponds to users familiar with the type of product, and "diversity" to users that have reviewed related products.