Predicting the emergence of self-radicalisation through social media: A complex systems approach

We are using a complex systems approach in a pilot project to create a data-driven real-time empirical analysis of the self-radicalisation phenomenon. The project, if successful, will generate actionable predictions about the likelihood that particular individuals of interest will become self-radicalised. In the pilot project, we use validated open source data to empirically test three hypotheses: individuals reveal their identity a personal signature through their texts; there exist tipping-point phenomena where identity may shift rapidly from one metastable state to another; and an individuals identity will show critical slowing down the characteristic dynamics that predict the approach of a tipping point.