A Survey on Forecasting Models for Preventing Terrorism

The security and welfare of any country are very crucial with states investing heavily to protect their territorial integrity from external aggression. However, the increase in the act of terrorism has given birth to a new form of security challenges. Terrorism has caused untold suffering and damages to civilian lives and properties and hence, finding a lasting solution to terrorism becomes inevitable. Until recently, the discourse on the nature and means of combating terrorism have largely been debated by politicians or statesmen. This study attempts an appraisal of machine learning survey on models in preventing terrorism via forecasting. Since the act of terrorism is dynamic in nature, i.e. their strategies change as counterterrorism methods are improved thereby requiring a more sophisticated way of predicting their moves. Some models discussed in this study include the Hawkes process, STONE, SNA, TGPM, and Dynamic Bayesian Network (DBN) which are all geared towards predicting the likelihood of a terrorist attack.