Research on Credit Scoring by Fusing Social Media Information in Online Peer-to-Peer Lending

In recent years, online Peer-to-Peer (P2P) lending market is rapidly expanding in China. In this paper, we use public dataset from PPDai, a leading online P2P platform in China to study the loan default. We construct a credit scoring model by fusing social media information based on decision tree. The experimental result shows that our model has good classification accuracy. From the credit scoring model and classification rules, we get a conclusion that the loan information, social media information, and credit information are most important factors for predicting the default. However, the credit rating is not as important as the platform described.