Visual investigation of similarities in Global Terrorism Database by means of synthetic social networks

During the last years, terrorist attacks over the world no longer can be considered as only sporadic accidents. This topic became an important problem and is solved as a global threat across several scientific disciplines. Terrorist attacks have been practiced by a wide array of organizations or groups for achieving their objectives. We can include political parties, nationalistic and religious groups, revolutionaries, ruling governments or others. Due to this fact the need of observing and discovering relations and rules of behavior based on terrorism incidents becomes very important. The authors of the paper present the usage of clustering methods and association rule mining methods for discovering and representation of potentially interesting similarities in the data. The purpose of the paper is to model synthetic social network based on relations obtained from the data about terroristic incidents to facilitate visual investigation of similarities in data and to study the network evolution during the years.