FCM-Based P2P Network Lending Platform Credit Risk Dynamic Assessment

Abstract

With the rapid development of Internet finance, Peer to Peer(P2P) network lending has also developed as a new type of financing. However, as the P2P network lending platform becomes more and more popular, problems such as running, shutting down, and withdrawing cash are emerging. One of the most common problems is caused by credit. This paper addresses the issues such as credit problems and existing risks of online platforms by proposing a Fuzzy Cognitive Map (FCM) that is used to establish a credit risk assessment model for the P2P network lending platform. It provides a new idea for the credit risk assessment of online lending platforms. The paper uses Lu Jinfu as an example to evaluate its credit risk and selects two other platforms to rank its credit risk. The results show that the fuzzy cognitive map considers the mutual influence and feedback between the indicators. The index weight matrix learned by historical data represents the relationship between the indicators. After inputting the initial data into the inference, the index data reaches a steady state. The steady-state refers to the state that we expect the indicator to reach after a certain period. As an objective evaluating method for the credit risk of the P2P network lending platform, the proposed method has demonstrated its advantage and effectiveness.