CUSTOMER CREDIT SCORING

Abstract

Credit risk analysis is a critical component of the banking and financial services industry, as it helps minimize losses caused by borrower defaults and ensures financial stability. Traditional methods of credit assessment are often subjective, inconsistent, and time-consuming, which can lead to inaccurate lending decisions. This project focuses on developing a machine learning-based system to predict the likelihood of a customer defaulting on a loan. The system utilizes a dataset containing demographic and financial information of loan applicants, including features such as age, education, employment years, income, and debt-related attributes. Data preprocessing techniques are applied to handle missing values and improve data quality. Additionally, new features such as total debt and debt ratio are engineered to enhance predictive performance. Exploratory Data Analysis (EDA) is performed to understand patterns and relationships within the data. A Random Forest Classifier is used to build the predictive model, achieving high accuracy and effectively capturing complex relationships between variables. Model performance is evaluated using metrics such as accuracy, confusion matrix, and classification report. The results highlight the importance of factors like debt ratio, total debt, and income in determining credit risk. Overall, this project demonstrates how machine learning can provide a reliable, objective, and scalable solution for credit scoring, improving loan approval processes and reducing financial risk