Credit Risk Assessment of P2P Lending Borrowers based on SVM

Abstract

With the development of Internet finance, peer to peer online (P2P) lending, which makes a win-win situation between lenders and borrowers, has become one of the most popular means of Internet finance in China. However, problem platforms and borrower default events have also occurred frequently with an explosive-speed growth of P2P online lending. Reducing credit risk of P2P lending borrowers still holds the key to the steady development of P2P online lending platforms. The results show that the SVM model based on cuckoo algorithm to optimize the parameter has a better classification accuracy. This model can be used to judge the potential credit risk of P2P lending borrowers and provides a theoretical basis for the risk management of Internet financial institutions at the same time.