Enhancing Loan Approval Accuracy Through Machine Learning and Behavioral Data Analysis

This research presents a machine learning (ML)-based system aimed at enhancing the process of personal loan approval in the banking sector. It addresses the limitations of traditional methods that rely heavily on manual verification and rule-based decision-making. The proposed system utilizes a dataset of 5,009 customers, incorporating demographic, financial, and behavioral features.The methodology includes data cleaning, feature selection, and the application of classification algorithms, as well as ensemble methods. Among the evaluated models, gradient boosting achieved an accuracy of 91.4%, while decision trees outperformed it with an accuracy of 97.92%, validating the system's robustness and scalability for large-scale applications.