Towards a Multidimensional Model for Terrorist Attacks Analysis and Mining

Terrorist attacks are in exponentially growth and cause daily dire human and material damages. Analyzing potential terrorism data is without doubt an indispensable task not only to understand these terrorist events, mainly terrorist implicated actors, their communities and their operation methods and tactics, but also to predict future attacks. However, terrorist events are commonly conducted differently and rarely fit the models of conventional attacks. Thus, terrorist events are usually hard to analyze, mine and investigate. In this paper, we propose a multidimensional model, named Terrorist Data Warehouse (TerDW), for allowing terrorist investigators to perform interesting and specific analysis for decision-making objectives. Practically, while consulting gathered information relative to specific terrorist attacks, such as attacks' locations, dates, used weapons, and implicated attackers, investigators can query the proposed data warehouse (TerDW) using multidimensional queries. The proposed model is based on consistent dimensions and measures. Terrorist attacks data are extracted from Global Terrorism Database (GTD), a well-known database that includes information about terrorist events that took place from 1970 to 2017. Examples of queries have been conducted to evaluate the efficiency of the proposed model. The queries' results demonstrate clearly that the proposed model allows investigators to carry out more appropriate multidimensional analysis, investigations and decisions.