Cross domain recommendation using multidimensional tensor factorization
作者：
Highlights：
• CD-MDTF Approach is proposed to alleviate the degree of sparsity and cold start.
• Results validated that recommendation accuracy is improved due to cross-domain.
• Tensor Factorization makes the modeling of domains as dimensions more flexible.
• Crawler is created to extract book genres corresponding to ISBN of books on Amazon.
• This paper provides future research directions in cross-domain recommendations.
摘要
•CD-MDTF Approach is proposed to alleviate the degree of sparsity and cold start.•Results validated that recommendation accuracy is improved due to cross-domain.•Tensor Factorization makes the modeling of domains as dimensions more flexible.•Crawler is created to extract book genres corresponding to ISBN of books on Amazon.•This paper provides future research directions in cross-domain recommendations.
论文关键词：Collaborative filtering,Cross domain recommendation systems,Tensor factorization,Clustering,Cross-domain multi-dimension tensor factorization
论文评审过程：Received 21 February 2017, Revised 14 September 2017, Accepted 15 September 2017, Available online 21 September 2017, Version of Record 13 October 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.09.042
