Skip to main content

Mining Pro-ISIS Radicalisation Signals from Social Media Users


Authors: Matthew Rowe & Hassan Saif

Date of Publication: 2016

Journal / Publisher: Proceedings of the Tenth International AAAI Conference on Web and Social Media


Purpose of the study

Key questions 

How can we detect when a user has adopted a pro-ISIS stance (i.e. is exhibiting radicalised behaviour on social media)?

What happens to Twitter users before they exhibit radicalized behaviour, and also after such exhibition?

What influences users to adopt pro-ISIS language?



Collect and study language in posts by Pro-ISIS, Anti-ISIS and Neutral content.

Number of participants

157,000 (users and followers in Europe)



Type of ‘participant’ 

Online users who create or share violent extremist content


Both mobilisation to fight abroad and mobilisation to conduct attacks on home soil.

Key findings 

Using content based and sharing behavior features, it is possible to identify “activation” points of social media users.

Key recommendations 

Identify if “activation” is associated with offline behaviors.

You might also like: