Detecting inorganic financial campaigns on Twitter

作者：

Highlights：

• The more viral a financial discussion is, the more social bots participate in it.

• The best model for detecting inorganic financial discussions achieved F1 = 0.97.

• Detection on the extent of inorganic financial content achieved R2 = 0.81.

• User, discussion, and financial-derived features proved to be highly informative.

摘要

•The more viral a financial discussion is, the more social bots participate in it.•The best model for detecting inorganic financial discussions achieved F1 = 0.97.•Detection on the extent of inorganic financial content achieved R2 = 0.81.•User, discussion, and financial-derived features proved to be highly informative.

论文关键词：Disinformation,Financial spam,Social bots,Twitter

论文评审过程：Received 15 September 2020, Revised 14 January 2021, Accepted 14 March 2021, Available online 24 March 2021, Version of Record 27 September 2021.

论文官网地址：https://doi.org/10.1016/j.is.2021.101769