Detecting terrorist influencers using reciprocal human-machine learning: The case of militant Jihadist Da’wa on the Darknet

Abstract

Abstract Over the past decade, social media has significantly impacted terrorism and counterterrorism, serving as a platform for incitement to violence under the guise of religious preaching. This study explores the critical role of preachers preaching righteous behavior, a process known as Da’wa in Islam. Focusing on Militant Jihadist Da’wa calling to violence, the research analyzes 6000 posts from Darknet forums associated with Jihadist groups from 2017 to 2018. The study improves the detection and understanding of militant Jihadist preaching by using an advanced method called Reciprocal Human-Machine Learning. The study demonstrates the feasibility of better detection and a deeper understanding of influencing terrorists.