Recommending the long tail items through personalized diversification
作者：
Highlights：
• If recommender systems focus on accuracy, they fail to recommend the long tail items.
• Thus, aspects such as diversity should be considered beside the accuracy.
• Different users may prefer different levels of diversity in the recommendations.
• To reduce the long tail problem, we personalize the diversification of recommendations.
• Our method alleviates the long tail problem without noticeable reduction in accuracy.
摘要
•If recommender systems focus on accuracy, they fail to recommend the long tail items.•Thus, aspects such as diversity should be considered beside the accuracy.•Different users may prefer different levels of diversity in the recommendations.•To reduce the long tail problem, we personalize the diversification of recommendations.•Our method alleviates the long tail problem without noticeable reduction in accuracy.
论文关键词：Recommender system,Diversity,Personalization,Long tail items,Multi-objective optimization,Simulated annealing
论文评审过程：Received 20 March 2018, Revised 1 November 2018, Accepted 2 November 2018, Available online 16 November 2018, Version of Record 19 December 2018.
论文官网地址：https://doi.org/10.1016/j.knosys.2018.11.004
