Conflict forecasting in the middle east with machine learning

This thesis aims to analyze and predict armed conflicts in the Middle East after 1970s by utilizing machine learning techniques, with a particular focus on logistic regression. The study involves statistically modeling the economic, political, and social factors that increase the likelihood of conflict through the use of supervised learning methods and statistical tests. By comparing data from periods with and without conflict, the research seeks to identify significant relationships between the dependent variable, namely the presence of conflict, and various independent variables. Through these modeling techniques, the study not only aims to explain the underlying dynamics of past conflicts but also to provide scientifically grounded predictions for potential future conflict zones. This research aspires to contribute to data-driven policymaking and to lay the groundwork for the development of early warning systems that promote peace and stability in fragile regions such as the Middle East. Keywords: Forecasting, Conflict, Middle East, Machine Learning