The five W’s of “bullying” on Twitter: Who, What, Why, Where, and When

作者：

Highlights：

• Machine learning is used to describe how bullying is represented on social media.

• More than 9 million posts referred to bullying episodes on Twitter over two years.

• Using social media data allows for an examination of all bullying role-players.

• Investigating real-time posts provides timely, dynamic data about bullying.

摘要

•Machine learning is used to describe how bullying is represented on social media.•More than 9 million posts referred to bullying episodes on Twitter over two years.•Using social media data allows for an examination of all bullying role-players.•Investigating real-time posts provides timely, dynamic data about bullying.

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

论文评审过程：Available online 12 December 2014.

论文官网地址：https://doi.org/10.1016/j.chb.2014.11.052