Balancing accuracy and diversity in recommendations using matrix completion framework
作者：
Highlights：
• Unified framework for joint accuracy-diversity optimization in recommender systems.
• Convex formulation utilizing item metadata for accuracy-diversity trade-off.
• Design an efficient algorithm using split Bregman technique for our formulation.
摘要
•Unified framework for joint accuracy-diversity optimization in recommender systems.•Convex formulation utilizing item metadata for accuracy-diversity trade-off.•Design an efficient algorithm using split Bregman technique for our formulation.
论文关键词：Recommender system,Matrix completion,Diversity,Metadata
论文评审过程：Received 9 June 2016, Revised 7 November 2016, Accepted 29 March 2017, Available online 30 March 2017, Version of Record 21 April 2017.
论文官网地址：https://doi.org/10.1016/j.knosys.2017.03.023
