Title: Predict personal customer credit risk using Linear regression algorithm for AI program
Authors: Trinh Thanh Do and Nguyen Ha Chi
Volume: 8
Issue: 11
Pages: 25-30
Publication Date: 2024/11/28
Abstract:
As financial institutions increasingly seek to manage credit risk effectively, the use of predictive models leveraging artificial intelligence (AI) and machine learning has gained significant traction. This report examines the implementation of a Linear Regression model to predict individual customer credit risk, incorporating key features such as age, income, loan amount, and credit score. The findings indicate that while the Linear Regression model can conduct basic analyses related to the correlation between these variables and credit risk, its limitations become apparent in situations involving non-linear relationships. To assess the accuracy of the model, performance metrics such as accuracy and F1 score were used. Furthermore, the report provides recommendations for improvement, including the application of Regularization, innovative feature generation techniques, and exploring other non-linear models to increase predictive performance.