Personalized Recommendation via Multi-dimensional Meta-paths Temporal Graph Probabilistic Spreading
作者：
Highlights：
• ●We propose a personalized recommendation method via multi-dimensional meta-paths temporal graph probabilistic spreading (MD-MP-TGPS).●We construct temporal multi-dimensional graphs, taking the interest drift of users, obsolescence and popularity of items, and dynamic update of interaction behavior data into full consideration.●We propose a dimension-free temporal graph probabilistic spreading framework via multi-dimensional meta-paths.●We propose two personalized recommendation strategies to automatically learn the priority and importance of these multi-dimensional meta-paths at the user-level granularity.●Comprehensive experiments show that the proposed MD-MP-TGPS method transcends the compared state-of-the-art methods in accuracy, diversity, and novelty of recommendation.
摘要
●We propose a personalized recommendation method via multi-dimensional meta-paths temporal graph probabilistic spreading (MD-MP-TGPS).●We construct temporal multi-dimensional graphs, taking the interest drift of users, obsolescence and popularity of items, and dynamic update of interaction behavior data into full consideration.●We propose a dimension-free temporal graph probabilistic spreading framework via multi-dimensional meta-paths.●We propose two personalized recommendation strategies to automatically learn the priority and importance of these multi-dimensional meta-paths at the user-level granularity.●Comprehensive experiments show that the proposed MD-MP-TGPS method transcends the compared state-of-the-art methods in accuracy, diversity, and novelty of recommendation.
论文关键词：Personalized recommendation,Meta-path,Probabilistic spreading,Temporal graph,Boosting strategy
论文评审过程：Received 24 April 2021, Revised 5 October 2021, Accepted 10 October 2021, Available online 18 October 2021, Version of Record 18 October 2021.
论文官网地址：https://doi.org/10.1016/j.ipm.2021.102787
