Developing a socio-computational approach to examine toxicity propagation and regulation in COVID-19 discourse on YouTube
作者：
Highlights：
• Topic modeling and social network analysis helped identify common themes and top toxic commenters and toxicity brokers on YouTube.
• Study shows how toxicity traverses within a network and affects non-toxic commenters.
• Commenter were found to be segregated based on toxicity.
• Commenter groups worked collectively to form echo-chambers to amplify toxic beliefs, some even displaying bot-like characteristics.
• Toxicity brokers were identified whose simulated removal reduced the overall toxicity of the network.
摘要
•Topic modeling and social network analysis helped identify common themes and top toxic commenters and toxicity brokers on YouTube.•Study shows how toxicity traverses within a network and affects non-toxic commenters.•Commenter were found to be segregated based on toxicity.•Commenter groups worked collectively to form echo-chambers to amplify toxic beliefs, some even displaying bot-like characteristics.•Toxicity brokers were identified whose simulated removal reduced the overall toxicity of the network.
论文关键词：Toxicity analysis,Social network analysis,Topic modeling,Pandemic,COVID-19,YouTube,Social media
论文评审过程：Received 22 January 2021, Revised 12 May 2021, Accepted 8 June 2021, Available online 10 June 2021, Version of Record 18 June 2021.
论文官网地址：https://doi.org/10.1016/j.ipm.2021.102660
