Pareto Pairwise Ranking for Fairness Enhancement of Recommender Systems

TL;DR

This paper introduces Pareto Pairwise Ranking, a new learning to rank algorithm that balances accuracy and fairness in recommender systems, demonstrating superior fairness performance while maintaining competitive accuracy.

Contribution

The paper presents a novel Pareto-based learning to rank algorithm that explicitly enhances fairness in recommender systems, addressing a gap in existing accuracy-focused methods.

Findings

Pareto Pairwise Ranking achieves competitive accuracy metrics.

It outperforms 9 other algorithms in fairness evaluation.

The method effectively balances fairness and accuracy.

Abstract

Learning to rank is an effective recommendation approach since its introduction around 2010. Famous algorithms such as Bayesian Personalized Ranking and Collaborative Less is More Filtering have left deep impact in both academia and industry. However, most learning to rank approaches focus on improving technical accuracy metrics such as AUC, MRR and NDCG. Other evaluation metrics of recommender systems like fairness have been largely overlooked until in recent years. In this paper, we propose a new learning to rank algorithm named Pareto Pairwise Ranking. We are inspired by the idea of Bayesian Personalized Ranking and power law distribution. We show that our algorithm is competitive with other algorithms when evaluated on technical accuracy metrics. What is more important, in our experiment section we demonstrate that Pareto Pairwise Ranking is the most fair algorithm in comparison with 9 other contemporary algorithms.