PREDICTING THE RISK LEVEL OF A LOAN BASED ON THE CUSTOMER'S PERSONAL FACTORS USING MACHINE LEARNING

ABSTRACT

Credit risk prediction in loan applications is a critical aspect of financial stability for both lenders and borrowers. This project, titled "Predicting the Risk Level of a Loan Based on the Customer's Personal Factors Using Machine Learning", focuses on assessing whether a loan carries potential credit risk or not. Many individuals face challenges in repaying loans due to financial mismanagement, leading to credit-related issues. To address this, the project aims to develop a machine learning-based solution that predicts the likelihood of credit risk based on customers’ personal and financial factors.

The primary objective is to help individuals and financial institutions make informed decisions by determining the credit amount a borrower can feasibly repay. By identifying potential risks beforehand, this application promotes financial awareness and assists in avoiding credit defaults, contributing to improved financial health for borrowers.

A variety of machine learning algorithms are employed, including Decision Tree, Random Forest, Support Vector Machine (SVM), Multi-layer Perceptron (MLP), Naive Bayes, and a stacking ensemble approach to enhance predictive accuracy. These models analyze diverse personal and financial attributes to classify loans as either "credit risk present" or "no credit risk present."

The project’s results are expected to provide a robust, user-friendly tool for evaluating creditworthiness, thereby benefiting a broad audience, including individuals, financial advisors, and lending institutions. By fostering better credit management practices, this initiative seeks to mitigate the risks associated with loan defaults.

Keywords: Credit Risk, Loan Prediction, Machine Learning, Financial Stability, Decision Tree, Random Forest, SVM, MLP, Naive Bayes, Stacking Ensemble.