Dynamic commonsense knowledge fused method for Chinese implicit sentiment analysis
作者：
Highlights：
• A sentimental commonsense knowledge graph embedded multi-polarity orthogonal attention model is proposed to learn the implication of the implicit sentiment. The method helps introduce the external sentimental commonsense knowledge and extend the original literal meaning.
• We analyzed the effectiveness of different knowledge relations in the ConceptNet knowledge base in detail, and proposed a matching and filtering method to select useful knowledge tuples for Chinese implicit sentiment analysis.
• A medium-sized golden annotation Chinese sentimental knowledge base is constructed which can be helpful for further sentiment analysis research.
摘要
•A sentimental commonsense knowledge graph embedded multi-polarity orthogonal attention model is proposed to learn the implication of the implicit sentiment. The method helps introduce the external sentimental commonsense knowledge and extend the original literal meaning.•We analyzed the effectiveness of different knowledge relations in the ConceptNet knowledge base in detail, and proposed a matching and filtering method to select useful knowledge tuples for Chinese implicit sentiment analysis.•A medium-sized golden annotation Chinese sentimental knowledge base is constructed which can be helpful for further sentiment analysis research.
论文关键词：Chinese implicit sentiment analysis,Sentimental knowledge,Knowledge fusion,Multi-polarity orthogonal attention
论文评审过程：Received 29 November 2021, Revised 10 February 2022, Accepted 20 March 2022, Available online 4 April 2022, Version of Record 4 April 2022.
论文官网地址：https://doi.org/10.1016/j.ipm.2022.102934
