Predicting the trend of compulsory earthquake insurance in Turkey by machine learning algorithms

This study aims to predict the tendency of homeowners in Turkey to purchase Compulsory Earthquake Insurance (CEI) using nine different machine learning (ML) classification algorithms. The dataset includes 11 independent variables related to demographic, structural, and geographic attributes, and one binary dependent variable. Data were collected via an online survey and preprocessed by removing incomplete responses. The Gradient Boosting (GB) algorithm achieved the highest performance with 70% accuracy. According to the SHAP (SHapley Additive exPlanations) analysis, "Monthly Income," "Square Meter Area," "Number of Floors," and "Construction Year" were the most significant features. The Odds Ratios (ORs) demonstrate that individuals with higher income and those living in newer and larger residences exhibit a significantly greater likelihood of possessing CEI. It was observed that individuals with higher education levels were less inclined to purchase insurance. Detailed analysis of the obtained results is expected to contribute to more effective policy-making in the insurance sector and among public authorities.