Boosting collaborative filtering with an ensemble of co-trained recommenders

作者：

Highlights：

• An ensemble-based co-training approach, named ECoRec, is proposed.

• ECoRec process data from two or more different views to create a more robust model.

• Our approach provide an enriched matrix that alleviate the sparsity and cold-start problems.

• Results show that our strategy improves the overall system’s performance.

摘要

•An ensemble-based co-training approach, named ECoRec, is proposed.•ECoRec process data from two or more different views to create a more robust model.•Our approach provide an enriched matrix that alleviate the sparsity and cold-start problems.•Results show that our strategy improves the overall system’s performance.

论文关键词：Co-training,Ensembles,Recommender systems,Semi-supervised learning

论文评审过程：Received 14 December 2017, Revised 18 July 2018, Accepted 11 August 2018, Available online 13 August 2018, Version of Record 23 August 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.08.020