Study On The Personal Credit Assessment Method For Online P2P Lending

P2P lending is a new type of loan mode formed by the intersection of Internet and traditional finance. It provides a more convenient loan platform, and brings new financing channels for individual, small and micro enterprise owners. In recent years,P2P industry in China is developing rapidly, but the phenomenon of collapse is getting worse, P2P loans is facing default risk and bad debt losses seriously. Credit evaluation is an important basis for managing loan default risk and supporting lending decision.However, compared with traditional loans, the financial data collected by P2P platform is limited, but the data type is more complicated, which also involves unstructured text information, causing new challenges for credit evaluation methods. Therefore, it is of great theoretical and practical significance to study the personal credit evaluation method for network lending in view of the characteristics of P2P lending.In this paper, research status of P2P loan and personal credit evaluation are first introduced. We analyze the characteristics of P2P loan mode and the problems of P2P credit evaluation. On the basis of this, the text analysis method and LDA model are used to extract the effective credit features from unstructured text information collected by the P2P platform, and the influence between text features and loan default has been validated through an empirical analysis and modeling method. Secondly, based on the architecture of combination model, a of two-stage feature selection method is designed to filter credit features in P2P platform,combined with a variety of measurement criteria.On this basis, the Random Forest method is used to construct the credit evaluation model for P2P lending. Finally,based on the data from a China's P2P platform, an experimental research is conducted to verify the effectiveness of credit features and credit evaluation methods we proposed in P2P lending.The results show that the text features extracted from unstructured information can improve the recognition rate of loan default and the accuracy of credit evaluation, and can be used as the basis of P2P credit evaluation. The feature combination selection method proposed in this paper and the credit evaluation model based on Random Forest have achieved good practical effect, which has certain reference significance for the credit evaluation of P2P network lending.