Hierarchical attentive knowledge graph embedding for personalized recommendation

作者：

Highlights：

• Exploit user-item connectivities in Knowledge Graphs for enhanced recommendation.

• Propose a novel framework to encode heterogeneous subgraphs between entities.

• Extensive experiments show the effectiveness of the proposed framework.

摘要

•Exploit user-item connectivities in Knowledge Graphs for enhanced recommendation.•Propose a novel framework to encode heterogeneous subgraphs between entities.•Extensive experiments show the effectiveness of the proposed framework.

论文关键词：Knowledge graphs,Graph neural network,Attention mechanism,Collaborative filtering,Recommender systems

论文评审过程：Received 4 March 2021, Revised 7 June 2021, Accepted 25 June 2021, Available online 30 June 2021, Version of Record 12 July 2021.

论文官网地址：https://doi.org/10.1016/j.elerap.2021.101071