Applying hybrid machine learning algorithms to assess customer risk-adjusted revenue in the financial industry

作者：

Highlights：

• Machine Learning methods are applied to predict customers’ RAR in P2P lending.

• Hybrid ML tools of supervised and unsupervised learning are used to predict RAR.

• Hybrid ML models present better accuracy and processing time than individual methods.

• Hybrid tools help identify relevant features for predicting RAR at the cluster level.

摘要

•Machine Learning methods are applied to predict customers’ RAR in P2P lending.•Hybrid ML tools of supervised and unsupervised learning are used to predict RAR.•Hybrid ML models present better accuracy and processing time than individual methods.•Hybrid tools help identify relevant features for predicting RAR at the cluster level.

论文关键词：P2P,Customer value prediction,Machine learning,Hybrid frameworks,Risk-adjusted revenue

论文评审过程：Received 8 December 2021, Revised 24 August 2022, Accepted 15 September 2022, Available online 20 September 2022, Version of Record 12 October 2022.

论文官网地址：https://doi.org/10.1016/j.elerap.2022.101202