Improving memory-based user collaborative filtering with evolutionary multi-objective optimization
作者：
Highlights：
• Memory-based collaborative filtering method is improved through a genetic algorithm.
• Multi-objective optimization is applied to find the appropriate group of neighbors.
• We propose an encoding that allows considering all possible neighbors.
• Our approach ensures accuracy and diversity of recommendations.
• We show the efficiency of our approach on benchmark and real-world datasets.
摘要
•Memory-based collaborative filtering method is improved through a genetic algorithm.•Multi-objective optimization is applied to find the appropriate group of neighbors.•We propose an encoding that allows considering all possible neighbors.•Our approach ensures accuracy and diversity of recommendations.•We show the efficiency of our approach on benchmark and real-world datasets.
论文关键词：Recommender systems,Collaborative filtering,Genetic algorithms,Multi-objective optimization,Diversity
论文评审过程：Received 23 September 2017, Revised 3 January 2018, Accepted 11 January 2018, Available online 12 January 2018, Version of Record 20 January 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.01.015
