DeepInteract: Multi-view features interactive learning for sequential recommendation
作者:
Highlights:
• Multi-view feature interactive learning was introduced for sequential recommendation.
• Interactive feature learning balanced the contradiction between static and dynamic features.
• DeepInteract was proposed to improve performance via multi-view feature learning.
• Interactive features were demonstrated to be important for sequential recommendation.
• DeepInteract outperformed state-of-the-art deep models significantly.
摘要
•Multi-view feature interactive learning was introduced for sequential recommendation.•Interactive feature learning balanced the contradiction between static and dynamic features.•DeepInteract was proposed to improve performance via multi-view feature learning.•Interactive features were demonstrated to be important for sequential recommendation.•DeepInteract outperformed state-of-the-art deep models significantly.
论文关键词:Recommender system,Multi-view feature interaction,Sequential recommendation,Attention network,Deep learning
论文评审过程:Received 28 May 2021, Revised 20 April 2022, Accepted 23 April 2022, Available online 30 April 2022, Version of Record 26 May 2022.
论文官网地址:https://doi.org/10.1016/j.eswa.2022.117305