Embedding Cultural Diversity in Prototype-based Recommender Systems

TL;DR

This paper proposes methods to reduce cultural popularity bias in prototype-based recommender systems by refining embedding spaces, leading to fairer and more inclusive recommendations without sacrificing accuracy.

Contribution

It introduces filtering and regularization techniques to mitigate demographic biases in prototype-based matrix factorization, enhancing fairness and diversity.

Findings

27% reduction in long-tail item rank

2% reduction in underrepresented country item rank

2% improvement in HitRatio@10

Abstract

Popularity bias in recommender systems can increase cultural overrepresentation by favoring norms from dominant cultures and marginalizing underrepresented groups. This issue is critical for platforms offering cultural products, as they influence consumption patterns and human perceptions. In this work, we address popularity bias by identifying demographic biases within prototype-based matrix factorization methods. Using the country of origin as a proxy for cultural identity, we link this demographic attribute to popularity bias by refining the embedding space learning process. First, we propose filtering out irrelevant prototypes to improve representativity. Second, we introduce a regularization technique to enforce a uniform distribution of prototypes within the embedding space. Across four datasets, our results demonstrate a 27\% reduction in the average rank of long-tail items and a 2\% reduction in the average rank of items from underrepresented countries. Additionally, our model achieves a 2\% improvement in HitRatio@10 compared to the state-of-the-art, highlighting that fairness is enhanced without compromising recommendation quality. Moreover, the distribution of prototypes leads to more inclusive explanations by better aligning items with diverse prototypes.