Local-to-global GCN with knowledge-aware representation for distantly supervised relation extraction

作者：

Highlights：

• A knowledge-aware framework is proposed to enhance the word representations.

• A piecewise attention method is used to distinguish the local and global information.

• A heterogeneous graph structure is designed to represent and encode sentence bags.

摘要

•A knowledge-aware framework is proposed to enhance the word representations.•A piecewise attention method is used to distinguish the local and global information.•A heterogeneous graph structure is designed to represent and encode sentence bags.

论文关键词：Relation extraction,Knowledge graph,Self-attention,Graph convolutional network

论文评审过程：Received 4 February 2021, Revised 30 September 2021, Accepted 2 October 2021, Available online 6 October 2021, Version of Record 25 October 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107565