Developing a hybrid intelligent approach to recognize terrorism / Ghada Mohammed Abdelaty Soliman ; Supervised Mohamed Hassan Rasmy , Omar Soliman Soliman , Motaz Mohammed Hosny Khorshid

Thesis (Ph.D.) - Cairo University - Faculty of Computers and Artificial Intelligence- Department of Operations Research and Decision Support

Terrorism is a complex adaptive system and it is one of the tactical phenomena that have been defeated for years and still none can guarantee, if it could be eliminated at all in the nearest future. Some governments indicated that terrorism does not happen in a vacuum, but rather happens for different social, economic, religious, psychological as well as political reasons.Terrorism has also been receiving closer attention from many researches in recent years. On the other side data mining is a new helpful technology that concentrates on dealing with information in order to make the best use of it.The recent machine learning paradigms play important roles to solve complex problems. The aim of this research is to propose a developed hybrid computational intelligent algorithm as a decision support (DS) tool for terrorism phenomenon. It combines the recent advanced machine learning paradigms and advanced data mining technology. It will be employed to predict, investigate the existence and networks growth of terrorism in different environments under different factors in order to predict the suspicious pattern of terrorism and so can be used to minimize the risk of this phenomenon.The proposed developed algorithm and hybrid framework can be applied on different environments to investigate and analyze the terrorist group (s) and then detect a well-known pattern of terrorism in order to be used as an early alarming tool to uncover future terrorist plots and so to avoid and minimize the terrorist attacks in the nearest future

Issued also as CD