Managing Popularity Bias in Recommender Systems with Personalized Re-ranking

Abstract

Many recommender systems suffer from popularity bias: popular items are recommended frequently while less popular, niche products, are recommended rarely or not at all. However, recommending the ignored products in the `long tail' is critical for businesses as they are less likely to be discovered. In this paper, we introduce a personalized diversification re-ranking approach to increase the representation of less popular items in recommendations while maintaining acceptable recommendation accuracy. Our approach is a post-processing step that can be applied to the output of any recommender system. We show that our approach is capable of managing popularity bias more effectively, compared with an existing method based on regularization. We also examine both new and existing metrics to measure the coverage of long-tail items in the recommendation.

arXiv admin note: text overlap with arXiv:1802.05382

References in corpus (1)

• Personalizing Fairness-aware Re-ranking

Cited by in corpus (17)

• Causal Intervention for Leveraging Popularity Bias in Recommendation
• Beyond Personalization: Research Directions in Multistakeholder Recommendation
• User-oriented Fairness in Recommendation
• Personalized Counterfactual Fairness in Recommendation
• The Unfairness of Popularity Bias in Recommendation
• Towards Fair Recommendation in Two-Sided Platforms
• Towards Individual and Multistakeholder Fairness in Tourism Recommender Systems
• Multi-stakeholder Recommendation and its Connection to Multi-sided Fairness
• NISER: Normalized Item and Session Representations to Handle Popularity Bias
• Debiasing Recommendation with Personal Popularity
• Towards Fair Personalization by Avoiding Feedback Loops
• A Comparative Analysis of Bias Amplification in Graph Neural Network Approaches for Recommender Systems
• Unfair Exposure of Artists in Music Recommendation
• An Adaptive Boosting Technique to Mitigate Popularity Bias in Recommender System
• Bandit based centralized matching in two-sided markets for peer to peer lending
• Understanding the Effects of Adversarial Personalized Ranking Optimization Method on Recommendation Quality
• Quantifying the Effects of Recommendation Systems