Loan evaluation in P2P lending based on Random Forest optimized by genetic algorithm with profit score

作者：

Highlights：

• Random Forest optimized by genetic algorithm with profit score (RFoGAPS) is proposed for loan evaluation in peer-to-peer (P2P) lending.

• A new profit score is proposed to evaluate the performance of the loan evaluation model, considering the actual and potential returns and losses.

• RFoGAPS outperforms several benchmark methods on a real-world dataset from Lending Club platform.

• Based on the experimental results, some suggestions about loan evaluation in P2P lending are proposed.

摘要

•Random Forest optimized by genetic algorithm with profit score (RFoGAPS) is proposed for loan evaluation in peer-to-peer (P2P) lending.•A new profit score is proposed to evaluate the performance of the loan evaluation model, considering the actual and potential returns and losses.•RFoGAPS outperforms several benchmark methods on a real-world dataset from Lending Club platform.•Based on the experimental results, some suggestions about loan evaluation in P2P lending are proposed.

论文关键词：P2P lending,Loan evaluation,Random Forest,Genetic algorithm,Profit score

论文评审过程：Received 26 March 2018, Revised 13 July 2018, Accepted 13 October 2018, Available online 16 October 2018, Version of Record 22 October 2018.

论文官网地址：https://doi.org/10.1016/j.elerap.2018.10.004