An overlapping clustering approach for precision, diversity and novelty-aware recommendations
作者：
Highlights：
• Scalability is improved using an overlapped clustering.
• Clusters of diverse and similar members are identified using a genetic algorithm.
• Recommendation relevance and diversity trade-off is controlled using a new encoding.
• Our approach improves accuracy, coverage, and novelty of recommendations.
• Experimentations on real-world datasets.
摘要
•Scalability is improved using an overlapped clustering.•Clusters of diverse and similar members are identified using a genetic algorithm.•Recommendation relevance and diversity trade-off is controlled using a new encoding.•Our approach improves accuracy, coverage, and novelty of recommendations.•Experimentations on real-world datasets.
论文关键词：Collaborative filtering,Diversity,Novelty,Overlapping clustering,Genetic algorithms
论文评审过程：Received 25 October 2019, Revised 3 March 2021, Accepted 14 March 2021, Available online 23 March 2021, Version of Record 12 April 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.114917
