<![CDATA[Finance plays a vital role as one of the key elements necessary for sustaining life. Since financial stability is closely linked to overall well-being, many individuals resort to borrowing from financial institutions. As a result, the increasing number of loan applications has led to a rise in financial burdens and fund congestion within these institutions. To mitigate such risks, credit scoring has become an essential predictive approach widely adopted in financial institutions to evaluate customer creditworthiness. Through credit scoring, institutions can determine whether a customer is eligible to receive a loan. This study employs an open-source dataset obtained from Kaggle and follows the CRISP-DM methodology, which consists of six phases: Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. The research implements a classification approach by comparing two algorithms—Random Forest Regression and XGBoost. The results show that the Random Forest Regression model performs better, achieving the highest accuracy, recall, and precision, with an AUC value of 0.796 and a Coefficient of Variation (CV) of 0.712]]>
