Abstract
The risk of credit loan exists when the bank issues a loan to the borrower, because the borrower has no way to repay the amount or defaults, which exposes the financial institution to the risk of loss. This causes financial institutions to suffer from effects that affect their creditworthiness, loss of capital and increased management and collection of loans. Factors of such risk include, but are not limited to, the creditworthiness of the borrower, the repayment ability of the borrower, and fluctuations in interest rates. With the development of economy, the credit loan risk of financial industry increases, and the harm is greater. After comparing KNN (K Nearest Neighbors), random forest and logistic regression using machine learning methods, we found that the model established by eXtreme Gradient Boosting method (XGBoost) can more accurately identify the risk of credit loans based on the characteristics of the borrower. Our research found that while the XGBoost model has better accuracy, there are also some areas that need improvement.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call