Detecting Social Polarization and Radicalization

You are here

This paper presents a theoretical computerized system to detect social polarization and to estimate the related chances of violent radicalization. Existing technologies are analyzed to determine how they can be integrated into the proposed system to fulfil the authors’ objectives. The authors propose a scalable design to enable the integration and extension of future state-of-the-art techniques into the system. This new system, DETECT, is utilized to identify trends and assess risks of social polarization and violent radicalization based on automated processing of online open source information, combined with human intelligence. It models and analyzes the total causal environment behind social polarization, radicalization and violent extremism. This is meant to allow for the early detection of polarization and the accurate prediction of violent events.

This article will be of use to practitioners and researchers looking to detect online radicalization and extremism using emerging technologies on digital platforms. DETECT collects sufficient data to allow the system to build social network graphs for polarized groups and apply social network analysis to identify key players, the presence of different communities within a network, and to find their preferences, skills, and the technologies they may have available. DETECT uses a four-step process: 1). Acquire online data and normalize it; 2). Process the data in various ways to extract maximum information content; 3). Investigate the processed information for presence of identified indicators; 4). Construct social network graphs and uncover graph internals like key players, skills involved, or technologies used. The authors demonstrate the utility of DETECT, the technological components required for it to function and the extensions of existing technologies that are needed to bring the technology to fruition. This theoretical exercise is important for the study of radicalization as it posits some of the technological platforms that may in the future be utilized in singular, group, or integrated fashion, to detect radicalization early and pre-empt violent events by radical individuals and groups. 

Pir Abdul Rasool Qureshi, Nasrulah Memon, Uff Kock Wiil and Panagiotis Karampelas

April 2011