Tailored text augmentation for sentiment analysis

作者：

Highlights：

• A novel data augmentation algorithm for sentiment analysis.

• Word sampling and replacing based on discriminative power and relevance to sentiment.

• Application of the algorithm to sentiment analysis on polices against COVID-19.

摘要

•A novel data augmentation algorithm for sentiment analysis.•Word sampling and replacing based on discriminative power and relevance to sentiment.•Application of the algorithm to sentiment analysis on polices against COVID-19.

论文关键词：Sentiment analysis,Text augmentation

论文评审过程：Received 16 September 2021, Revised 13 January 2022, Accepted 15 May 2022, Available online 26 May 2022, Version of Record 14 June 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117605