Gated Tree-based Graph Attention Network (GTGAT) for medical knowledge graph reasoning

作者：

Highlights：

• A gated tree-based graph model (GTGAT) for distilling the neighbor representations.

• Multi-semantic contents are utilized for handling complex real-world knowledge graphs.

• Our approach shows better performance than the previous state-of-the-art for both the transductive and inductive reasoning.

摘要

•A gated tree-based graph model (GTGAT) for distilling the neighbor representations.•Multi-semantic contents are utilized for handling complex real-world knowledge graphs.•Our approach shows better performance than the previous state-of-the-art for both the transductive and inductive reasoning.

论文关键词：Medical knowledge graph,Graph attention network,Disease diagnosis

论文评审过程：Received 12 September 2021, Revised 28 April 2022, Accepted 29 May 2022, Available online 10 June 2022, Version of Record 21 June 2022.

论文官网地址：https://doi.org/10.1016/j.artmed.2022.102329