Research on Personal Loan Prediction Based on XGBoost and SHAP

This paper investigates the issue of personal loan prediction based on XGBoost and SHAP. Personal loans are a crucial part of the financial sector, holding significant importance for both individuals and financial institutions. However, personal loans carry certain risks and uncertainties, making accurate prediction of repayment ability critical for financial institutions. We delve into the application of XGBoost in personal loan prediction, covering aspects such as feature selection and extraction, model training, and evaluation. Subsequently, we introduce the concept of SHAP as an explainable machine learning technique, discussing its application in feature interpretation and its advantages in personal loan prediction. Finally, we present a case study on personal loan prediction based on SHAP, demonstrating the effectiveness and feasibility of this approach through experiments. The results of this research are of great significance for improving the accuracy and stability of personal loan predictions.