Prediction of Loan Overdue Based On Machine Learning Algorithm

Abstract

With the development of the financial Internet, analyzing the repayment ability and willingness of loan objects has become a key link. This paper uses the loan overdue data set of the TianChi platform to predict whether the borrower is in default or not. Firstly, Feature engineering is used to get the useful features for training. Then this paper compares the performance of five machine learning algorithms on predicting the loan overdue. The results show that the LightGBM model has the best performance and stability.