Credit risk evaluation model with textual features from loan descriptions for P2P lending

作者：

Highlights：

• A P2P lending credit risk model with textual descriptions taking into consideration is proposed.

• The proposed model improves the accuracy of default prediction, which helps not only platforms in their loan approval process, but also investors in their investment decision making.

• This work extends the application of deep learning and text mining related techniques in financial realm.

摘要

•A P2P lending credit risk model with textual descriptions taking into consideration is proposed.•The proposed model improves the accuracy of default prediction, which helps not only platforms in their loan approval process, but also investors in their investment decision making.•This work extends the application of deep learning and text mining related techniques in financial realm.

论文关键词：Peer-to-peer lending,Credit risk model,Transformer encoder,Loan description

论文评审过程：Received 9 December 2019, Revised 5 May 2020, Accepted 28 May 2020, Available online 8 July 2020, Version of Record 23 July 2020.

论文官网地址：https://doi.org/10.1016/j.elerap.2020.100989