Managing Popularity Bias in Recommender Systems with Personalized Re-ranking
By Himan Abdollahpouri, Robin D. Burke, Bamshad Mobasher. The Florida AI Research Society, 2019
333 citations recorded by the source corpus.
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.
Subfields: Information Systems, Management Science and Operations Research, Marketing
View Paper: https://www.semanticscholar.org/paper/1fde04a2fcbd199c492a17d8c56d638676287193
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Information Systems
- Xiangnan He
- Fuli Feng
- Yongfeng Zhang
- Dietmar Jannach
- Hanxiong Chen
- Tianxin Wei
- Jiawei Chen
- Gediminas Adomavicius
In Management Science and Operations Research
- Xiangnan He
- Fuli Feng
- Dietmar Jannach
- Robin D. Burke
- Tianxin Wei
- Himan Abdollahpouri
- Jiawei Chen
- Yongdong Zhang
In Marketing
- Jiawei Chen
- Yingqiang Ge
- Yongdong Zhang
- Dietmar Jannach
- Yongfeng Zhang
- Gediminas Adomavicius
- Robin D. Burke
- Zuohui Fu
Citations in the corpus
- Model-Agnostic Counterfactual Reasoning for Eliminating Popularity Bias in Recommender System (2020, background): “[3] introduces a re-ranking approach that can be applied to the output of the recommender systems.”
- Causal Intervention for Leveraging Popularity Bias in Recommendation (2021, methodology): “- xQuAD [3].”
- Multistakeholder recommendation: Survey and research directions (2020, background): “The causes of unfair outcomes in recommendation have only been lightly studied, although Ekstrand et al. (2018a) and Abdollahpouri et al. (2019b) provide some important clues, namely that the size of different user groups is not always a good indicator of outcomes.”
- User-oriented Fairness in Recommendation (2021, background): “Such researches mainly solve this problem by increasing the number of recommended unpopular items (long-tail items) or otherwise the overall catalog coverage [2, 3, 5, 21].”