P2P Online Loan Credit Risk Identification Scheme Planning Based On Convolutional Neural Network Model

P2P online lending is a business model that gathers funds and lends them to people in need of funds.It borrows money through the Internet and it is committed to customers beyond the reach of traditional finance.Although the number of P2 P platforms has shown a trend of explosive growth in recent years,this does not mean that the P2 P industry is in a stage of healthy development.As a kind of credit business,P2 P online lending companies are generally small in scale and far less capable of risk management than commercial Banks,so the biggest risk they face is credit risk.This paper studies the personal credit risk assessment and prediction based on Lending Club transaction data.The loan default prediction based on Lending Club data can help the platform solve the problem of information asymmetry,provide certain reference opinions for investors,and provide certain reference opinions for China's P2 P credit risk assessment system.Although there are many evaluation methods for this problem,most of them are based on machine learning methods for evaluation and prediction.However,the prediction effect of these traditional machine learning methods is very dependent on the characteristics of artificial design,and the method of artificial design often fails to take into account all the characteristics,and at the same time,the artificial design features need to spend a lot of time and labor cost.In recent years,the deep learning method has gradually entered people's line of sight and received more and more attention.Convolution Neural Network(CNN)is one of the classic and wide application of Network structure.This paper attempts to apply this deep learning method to the measurement of credit risk of P2 P online loans,so as to achieve better prediction effect and serve P2 P platforms and platform investors.Therefore,this paper takes lending club website data as an example.Firstly,it analyzes the basic information of website users to get the basic portrait of users in breach of the agreement.Then,SMOTE method is used to solve the problem of unbalanced data set.The prediction results were compared with those obtained by other traditional machine learning methods.The results showed that the prediction accuracy of customer default risk based on the convolutional neural network model was significantly higher than that of other models,and the default probability of credit risk of platform borrowers could be evaluated more accurately.At the same time,convolutional neural network model can automatically learn features from data,which can save a lot of time compared with manual design features.Therefore,the model established in this paper has more advantages in the field of credit risk assessment in the Internet financial industry.In this paper,CNN model is applied to the identification of credit risk of P2 P online loan,and scene analysis is carried out.Convolutional neural network is applied to specific business scenarios to provide more efficient identification model and certain reference opinions for P2 P online loan platform and platform investors.At the same time,convolutional neural network and other deep learning algorithms can not only be applied to P2 P online lending platforms,but also provide certain reference opinions for consumer finance and other industries.