iz

Terrorist attacks are a major problem and a leading concern in the entire world. It is one of the most important topics for all government officials to consider. There are many basic connections and patterns in the data associated with terrorist attacks. If these potential phenomena can effectively control the development of early warning systems for terrorism, they can help to solve complex terrorist decisionmaking issues and maintain regular detection and predictions. Machine learning (ML) models have made significant contributions to the development of prediction models by simulating the dynamic statistical structures of terrorism. This has resulted in improved results and cost-effective strategies. Researchers hope to find more reliable and effective prediction models by incorporating new machine learning approaches and hybridizing current ones. This article aims to provide a concise review of the various machine learning algorithms used to predict terrorism using the Global Terrorism Database GTD historical data set by START. We could use this study as a reference for researchers when deciding which machine learning approach to use for the required prediction challenge.

Keywords: GTD, ML Algorithms, Terrorism Prediction, classification, clustering.