Study of radical views on social media: Classification and group dynamics analysis

Social media platforms have changed the way extremist groups recruit, influence and potentially radicalize users online. Radicalization has drawn the attention of many researchers worldwide with its increasing presence on social media. Therefore, the ability to identify and classify radical views online and their potential role in radicalizing individuals is of critical importance. In this work, we address this challenge by combining a number of machine learning and natural language processing techniques such as sentiment analysis, named entity recognition, clustering, opinion leader identification, and social circle analysis to gain insights about radical views related to extreme nationalism, and understand the social dynamics underpinning extremists' related interactions. The proposed approach was implemented and tested on more than 259, 000 tweets collected over a five-year span. The different analysis steps were presented and the results obtained were analyzed.