A scalable and robust trust-based nonnegative matrix factorization recommender using the alternating direction method
作者：
Highlights：
• A social regularization recommender system method called TrustANLF is proposed.
• TrustANLF incorporates the social trust information of users in the NMF framework.
• Trust statements are used to deal with the data sparsity and cold-start issues.
• Alternating direction optimization method is used to improve the convergence speed.
• The results of experiments reveal the effectiveness of the proposed method.
摘要
•A social regularization recommender system method called TrustANLF is proposed.•TrustANLF incorporates the social trust information of users in the NMF framework.•Trust statements are used to deal with the data sparsity and cold-start issues.•Alternating direction optimization method is used to improve the convergence speed.•The results of experiments reveal the effectiveness of the proposed method.
论文关键词：Recommender systems,Social trust,Matrix factorization,Alternating direction method,Collaborative filtering
论文评审过程：Received 9 February 2018, Revised 7 December 2018, Accepted 12 December 2018, Available online 14 December 2018, Version of Record 23 January 2019.
论文官网地址：https://doi.org/10.1016/j.knosys.2018.12.016
