Semantics perception and refinement network for aspect-based sentiment analysis

作者：

Highlights：

• Semantics perception and refinement network addresses the issues of current ABSA methods.

• Dual gated multichannel convolution acquires the aspect-related semantic features.

• Multichannel convolution extracts informative local semantic features.

• Dual refinement gate strengthens the interaction and noise filter.

摘要

•Semantics perception and refinement network addresses the issues of current ABSA methods.•Dual gated multichannel convolution acquires the aspect-related semantic features.•Multichannel convolution extracts informative local semantic features.•Dual refinement gate strengthens the interaction and noise filter.

论文关键词：Aspect-based sentiment analysis,Dual refinement gate,Multichannel convolution

论文评审过程：Received 29 September 2020, Revised 16 December 2020, Accepted 2 January 2021, Available online 7 January 2021, Version of Record 11 January 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.106755