Leveraging Machine Learning for Credit Scoring: An Evaluation of the CART Method in Assessing Housing Loan Eligibility

This study investigates the application of the Classification and Regression Tree (CART) method for evaluating housing loan eligibility in XYZ Cooperative, a financial institution in Jakarta.The research aims to address the challenges faced by the cooperative in conducting manual loan feasibility analyses, which have led to inefficiencies and prolonged application processes.Using historical loan data and relevant borrower variables, the study develops a credit scoring model based on the CART method to improve the accuracy and efficiency of predicting borrower feasibility.The results demonstrate that the CART-based model enhances the credit assessment process, reduces the risk of loan defaults, and accelerates decision-making.The study contributes to the literature by showcasing the effectiveness of the CART method in a cooperative lending context and provides practical recommendations for XYZ Cooperative to improve their credit risk management system.However, the generalizability of the findings beyond the specific case of XYZ Cooperative remains a limitation.Future research could extend this study by incorporating additional variables, comparing different credit scoring methods, and validating the model in other financial institutions.