The Research On Credit Risk Assessment Of P2P Network Loan Based On Fuzzy Neural Network

P2P(peer-to-peer)network lending is an innovative financial pattern that combines internet finance,petty loan and private lending.Borrowers and investors use e-commerce platform which built by e-commerce companies to finish peer-to-peer lending activities.P2P network lending has the advantages of low cost of entry and channel,flexible and efficient way of transaction,relatively transparent borrowing method,it is highly sought after by individuals and small and micro enterprises.Since entering our country in 2007,P2P network lending industry has experienced rapid development,China has become the world's largest P2P trading market at present.However,frequent "Foot" event and borrower default events reflect the high risk and high bad debt rate prevalent in P2P platform.High credit risk becomes the biggest bottleneck in the development of network loan.Based on credit risk,this paper analyzes the origin and characteristics of the credit risk of the borrowers in the P2P network,and constructs a credit risk evaluation index system and quantification standard based on the personal credit risk assessment method of the commercial banks,which is suitable for China's domestic P2P network loan borrowers.Then,we use the entropy method to determine the weight of the index.In the aspect of model selection,based on the theory and method of credit risk of P2P network lending borrowers,this paper analyzes the feasibility of the combination of fuzzy logic and BP neural network.By using the real sample data of domestic P2P platform,the model of fuzzy neural network credit risk assessment is constructed,and the data training and model simulation are carried out.Compared with BP neural network model,the results show that the accuracy of fuzzy neural network prediction is higher.The result of robustness test proves that the model of fuzzy neural network can judge the credit of the borrower with high precision and can provide the basis for the loan of P2P network.Based on the research of fuzzy neural network model,this paper puts forward some countermeasures and suggestions:expanding the application of fuzzy neural network in credit evaluation,improving the external system and environment of credit risk assessment of P2P network lending,accelerating the construction of internal control system of P2P network lending credit risk.