Bullying discourse on Twitter: An examination of bully-related tweets using supervised machine learning

作者：

Highlights：

• Twitter serves as a platform to share both online and offline bullying episodes.

• Most tweets from the victim's perspective and commonly to report or self-disclose.

• Bullying-related tweets were significantly longer than the average tweet.

• High profile bullying incidents prompted an increase in bullying-related tweets.

• The study included strategically expanded keywords with cyberbullying keywords.

摘要

•Twitter serves as a platform to share both online and offline bullying episodes.•Most tweets from the victim's perspective and commonly to report or self-disclose.•Bullying-related tweets were significantly longer than the average tweet.•High profile bullying incidents prompted an increase in bullying-related tweets.•The study included strategically expanded keywords with cyberbullying keywords.

论文关键词：Bullying,Twitter,Cyberbullying,Machine learning,Social media

论文评审过程：Received 9 June 2020, Revised 21 January 2021, Accepted 8 February 2021, Available online 10 February 2021, Version of Record 22 February 2021.

论文官网地址：https://doi.org/10.1016/j.chb.2021.106735