Modeling the Causes of Terrorism from Media News: An Innovative Framework Connecting Impactful Events with Terror Incidents

Terrorism has become an increasingly relevant issue accounting for significant social, economic and political impact. Due to powerful media coverage on the subject, a lot of information is now publicly available, although normally found in an unstructured form. This research aims to better understand the connection between a collection of impactful events, such as external or internal conflicts and military operations, with terror events and its motivations. To this end, a framework was devised, starting with an online news scraper, coupled with machine learning and natural language processing techniques, capable of clustering keywords into the main topics found in the news. The results of these algorithms, in the form of structured data, were later fed to a modeling technique capable of finding, to a certain degree, the connections between topics and terror events. The approach presented in this work adopts a perspective that, to the best of our knowledge, has not been previously seen in specialized literature. Furthermore, this methodology constitutes the groundwork for open source intelligence, capable of being applied to various similar domains like the prediction of political risk index or economic risk index. © 2018 IEEE.