Pareto Pairwise Ranking for Fairness Enhancement of Recommender Systems

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.

References in corpus (7)

• BPR: Bayesian Personalized Ranking from Implicit Feedback
• Controlling Fairness and Bias in Dynamic Learning-to-Rank
• ZeroMat: Solving Cold-start Problem of Recommender System with No Input Data
• PoissonMat: Remodeling Matrix Factorization using Poisson Distribution and Solving the Cold Start Problem without Input Data
• KL-Mat : Fair Recommender System via Information Geometry
• RankMat : Matrix Factorization with Calibrated Distributed Embedding and Fairness Enhancement
• Extremal GloVe: Theoretically Accurate Distributed Word Embedding by Tail Inference

Cited by in corpus (2)

• The Fallacy of Borda Count Method -- Why it is Useless with Group Intelligence and Shouldn't be Used with Big Data including Banking Customer Services
• LogitMat : Zeroshot Learning Algorithm for Recommender Systems without Transfer Learning or Pretrained Models