The language and targets of online trolling: A psycholinguistic approach for social cybersecurity

作者：

Highlights：

• Online trolling statistically has less complexity and more abusive targeted language.

• Psycholinguistic troll detection models compete with deep learning and transformers.

• Trolls stoke conflict in bipartisan audiences, while bots amplify right-wing media.

• We show the end-to-end value of theory-informed computing to address disinformation.

摘要

•Online trolling statistically has less complexity and more abusive targeted language.•Psycholinguistic troll detection models compete with deep learning and transformers.•Trolls stoke conflict in bipartisan audiences, while bots amplify right-wing media.•We show the end-to-end value of theory-informed computing to address disinformation.

论文关键词：Trolls,Bots,Disinformation,Social media,Machine learning,Psycholinguistics

论文评审过程：Received 18 March 2022, Revised 20 June 2022, Accepted 27 June 2022, Available online 8 July 2022, Version of Record 8 July 2022.

论文官网地址：https://doi.org/10.1016/j.ipm.2022.103012