To search, Click below search items.


All Published Papers Search Service


Improve Risk Prediction in Online Lending (P2P) Using Feature Selection and Deep Learning


Nguyen Truong Thang, Khuat Thanh Son, Ngo Thi Thu Trang, Nguyen Ha Nam, and Tran Manh Dong


Vol. 19  No. 11  pp. 216-222


At the beginning of the 21st century, Peer2Peer (P2P) lending was established and developed rapidly in the United Kingdom, the United States, China and some other countries. The main challenge for individual investors in the P2P lending market is to allocate their money effectively through various lendings by accurately assessing the risk level of each lending. Traditional scoring models cannot fit the needs of individual investors in P2P lending because they do not provide natural mechanisms for asset allocation since for P2P lendings there are no traditional financial institutions. In this study, the report proposes a new way to analyze data for this emerging market. We have designed a risk scoring model based on advanced machine learning methods, capable of assessing the profits and risks of personal lendings. The report also applies a feature selection method and removes extraneous features to improve the efficiency of machine learning models. We conducted experiments on real lending datasets from the P2P lending markets. Test results show that the proposed model can improve the investment efficiency compared to existing P2P lending scoring methods.


P2P, credit scoring, AI, deep learning, Boltzmann machine, feature selection, financial risk.