The project, called TRIVALENT, is looking at ways to protect citizens around the globe from being targeted, and potentially enticed into endorsing and propagating violent radical content. The goal is to create a more comprehensive understanding of violent radicalisation, by investigating the psychological and behavioral patterns, models and motivations for individuals. This will include the assessment of radicals’ language, integrations into online networks of radicals and the spread of radicalisation material.
Content mining techniques will be developed for analysing radicalisation narratives on websites and social media, and used together with statistical, linguistic, semantic, and social data analysis approaches to detect signs of online radicalisation.Data will be collected from social media, and models will be developed to automatically detect radicalised content and user accounts, and to distinguish violent from non-violent radicals. How people judge radical messages on social media will also be studied to determine any similarities and differences in these judgements. Computational methods will also be developed that can distinguish between content that supports radicalisation, from content that only reports on it.Further information can be found on the TRIVALENT project websitehttp://trivalent-project.eu/
Fernandez, Miriam; Asif, Moizzah and Alani, Harith (2018). Understanding the Roots of Radicalisation on Twitter. In: In WebSci ’18: 10th ACM Conference on Web Science, 27-30 May 2018, Amsterdam, Netherlands, ACM (Association for Computing Machinery). Saif, Hassan; Dickinson, Thomas; Kastler, Leon; Fernandez, Miriam and Alani, Harith (2017). A Semantic Graph-Based Approach for Radicalisation Detection on Social Media. In: ESWC 2017: The Semantic Web - Proceedings, Part I, Lecture Notes in Computer Science, pp. 571–587.