Clustering Terrorist Groups using Their Activity and Lethality Measures

Abstract:

Analyzing terrorist groups’ trends using innovative Data Analysis techniques was widely used in many research fields related to fighting terrorism. The obtained results aimed to enhance the capabilities of Homeland Security departments and intelligence services to better draw insights from the huge amount of data gathered. Data analysis models are used to identify crime trends, investigate unsolved attacks, predict potential threats, etc. One motivating issue, that was not widely developed in the literature, is the clustering of terrorist groups based on some adequate parameters to provide actionable information. In this study, the most well-known Islamist terrorist groups identified between 1970 and 2016 in North Africa are investigated These groups were partitioned based on their activity and lethality parameters that can provide highly valuable results concerning their spread and fame of these groups. This study was carried out using the open source Global Terrorism Database (GTD) dataset.