FaiRIR: Mitigating Exposure Bias from Related Item Recommendations in Two-Sided Platforms

TL;DR

This paper identifies skewed exposure distribution in related item recommendation systems and introduces FaiRIR interventions that improve fairness with minimal impact on recommendation quality.

Contribution

The paper proposes flexible interventions (FaiRIR) to mitigate exposure bias in RIR systems, applicable to existing algorithms like rating-SVD and item2vec.

Findings

Existing RIR algorithms cause skewed exposure distribution.

FaiRIR interventions enable fine-grained control of exposure.

Interventions maintain recommendation quality with minimal trade-offs.

Abstract

Related Item Recommendations (RIRs) are ubiquitous in most online platforms today, including e-commerce and content streaming sites. These recommendations not only help users compare items related to a given item, but also play a major role in bringing traffic to individual items, thus deciding the exposure that different items receive. With a growing number of people depending on such platforms to earn their livelihood, it is important to understand whether different items are receiving their desired exposure. To this end, our experiments on multiple real-world RIR datasets reveal that the existing RIR algorithms often result in very skewed exposure distribution of items, and the quality of items is not a plausible explanation for such skew in exposure. To mitigate this exposure bias, we introduce multiple flexible interventions (FaiRIR) in the RIR pipeline. We instantiate these mechanisms with two well-known algorithms for constructing related item recommendations -- rating-SVD and item2vec -- and show on real-world data that our mechanisms allow for a fine-grained control on the exposure distribution, often at a small or no cost in terms of recommendation quality, measured in terms of relatedness and user satisfaction.