Research On Network Loan Credit Risk Prediction Model Based On Machine Learning Algorithm

From the initial bank loan to the current P2 P network lending,we have to admit that Internet finance is changing the traditional lending industry,and even subverting people's lives.In the era of Internet plus finance,P2 P network lending can be said to be promising.Internet lending is born in this environment.It provides a transaction for both sides of the transaction,that is,borrowers and investors.Platform,which evaluates the credit degree of the borrower according to the basic information and borrowing information provided by the borrower,and then evaluates whether the borrower will default in the future.In this paper,the risk brought by the default of the borrower is called credit risk.Through analyzing and modeling the borrower's personal basic information and borrowing information,this paper predicts whether the borrower will default in the future.In some cases,three methods,decision tree,random forest and support vector machine,are used to predict the target state respectively,and the prediction results of the three methods are compared by the accuracy measurement results.The results show that among the three methods mentioned above,the random forest algorithm has the best prediction effect,and the method can select the variables that need to be focused on according to the importance of the variables.