Investment pattern clustering based on online P2P lending platform
Shen Fan, Nianlong Luo · 2016
With the rapid development of online peer-to-peer (P2P) lending platforms in recent years, more and more people participate in the borrowing and lending transactions online. The risk attitude of online lenders normally determines the size of capital invested on the platforms during different periods. Although investment time series from each lender is unique, they share similar characteristics of investment trends. Based on the data from PPDAI platform in China, this paper proposes an effective approach of data preprocessing, namely Key Points Approximate Fitting (KPAF) algorithm, to identify different investment patterns. The KPAF algorithm contributes to a better accuracy of clustering.