GL-GCN: Global and Local Dependency Guided Graph Convolutional Networks for aspect-based sentiment classification

作者：

Highlights：

• Exploit the syntactic dependency structure to mine sentence local structure information.

• Construct a word-document graph to explore global word dependency information.

• Propose a novel architecture to encode both global and local structure signals.

• Conduct extensive experiments to verify the effectiveness of the proposed approach.

摘要

•Exploit the syntactic dependency structure to mine sentence local structure information.•Construct a word-document graph to explore global word dependency information.•Propose a novel architecture to encode both global and local structure signals.•Conduct extensive experiments to verify the effectiveness of the proposed approach.

论文关键词：Graph convolutional networks,Aspect-based sentiment classification,Attention mechanism,Sentiment analysis

论文评审过程：Received 7 December 2020, Revised 2 June 2021, Accepted 1 August 2021, Available online 11 August 2021, Version of Record 16 August 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115712