Exploring user movie interest space: A deep learning based dynamic recommendation model
作者：
Highlights：
• A sequence-based deep learning model for movie recommendation is proposed.
• User Movie Interest Space (UMIS) is introduce to reflect user similarities.
• Three indexes are developed to describe UMIS from multiple points of interest.
• Rating sequence predictions based on UMIS perform better than those without UMIS.
• The method greatly improves dynamic recommendation performance than other methods.
摘要
•A sequence-based deep learning model for movie recommendation is proposed.•User Movie Interest Space (UMIS) is introduce to reflect user similarities.•Three indexes are developed to describe UMIS from multiple points of interest.•Rating sequence predictions based on UMIS perform better than those without UMIS.•The method greatly improves dynamic recommendation performance than other methods.
论文关键词：Intelligent recommendation systems,User movie interest space,Dynamic interest flow,Deep learning
论文评审过程：Received 21 December 2019, Revised 5 January 2021, Accepted 5 February 2021, Available online 11 February 2021, Version of Record 25 February 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.114695
