Diversity in news recommendations using contextual bandits
作者：
Highlights：
• We propose novel ‘socially responsible’ way of dynamic recommendation.
• Our algorithm is based on contextual bandit models.
• The algorithm is relevant for the cases when, MF based algorithms are not applicable.
• It results in a more balanced distribution and a diverse set of recommended articles.
• Experiments demonstrate the trade-off between clicks and diversity.
摘要
•We propose novel ‘socially responsible’ way of dynamic recommendation.•Our algorithm is based on contextual bandit models.•The algorithm is relevant for the cases when, MF based algorithms are not applicable.•It results in a more balanced distribution and a diverse set of recommended articles.•Experiments demonstrate the trade-off between clicks and diversity.
论文关键词：Recommender systems,Contextual bandit,Fairness,Equitable machine learning
论文评审过程：Received 24 April 2020, Revised 23 March 2021, Accepted 26 December 2021, Available online 15 January 2022, Version of Record 1 February 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.116478
