Novel automatic group identification approaches for group recommendation

作者：

Highlights：

• A bisecting clustering based approach for automatic group identification is proposed.

• The proposed clustering approach helps forming groups with restricted size.

• Item genres-based profiling mechanisms are adopted to improve group recommendations.

• Demographic features of users are utilized for increasing group homogeneity.

摘要

•A bisecting clustering based approach for automatic group identification is proposed.•The proposed clustering approach helps forming groups with restricted size.•Item genres-based profiling mechanisms are adopted to improve group recommendations.•Demographic features of users are utilized for increasing group homogeneity.

论文关键词：Group recommender systems,Bisecting k-means clustering,Automatic group identification,User-profiling

论文评审过程：Received 3 June 2020, Revised 3 January 2021, Accepted 9 February 2021, Available online 17 February 2021, Version of Record 15 March 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114709