Flash mobs, Arab Spring and protest movements: Can we analyse group identities in online conversations?

作者：

Highlights：

• We present a model to analyse online group identities on textual conversations.

• Our model applies data mining, NLP and sylometric techniques.

• Our model detects the salience of group identities with 95% accuracy.

• Our model is able to distinguish group identities from others with 84% accuracy.

• We identify features that may enable mal-actors to manipulate online groups.

摘要

•We present a model to analyse online group identities on textual conversations.•Our model applies data mining, NLP and sylometric techniques.•Our model detects the salience of group identities with 95% accuracy.•Our model is able to distinguish group identities from others with 84% accuracy.•We identify features that may enable mal-actors to manipulate online groups.

论文关键词：Social identities,Online social media,Natural language processing

论文评审过程：Received 10 February 2016, Revised 11 April 2016, Accepted 12 June 2016, Available online 16 June 2016, Version of Record 23 June 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2016.06.023