Rule-based credit risk assessment model using multi-objective evolutionary algorithms

作者：

Highlights：

• This study considered the generation of classification rule as an optimization problem.

• We present a comparative study of four multi-objective evolutionary algorithms.

• The used algorithms provide a better trade-off between accuracy and interpretability.

• The proposed a model aims to select the most relevant attributes for decision.

摘要

•This study considered the generation of classification rule as an optimization problem.•We present a comparative study of four multi-objective evolutionary algorithms.•The used algorithms provide a better trade-off between accuracy and interpretability.•The proposed a model aims to select the most relevant attributes for decision.

论文关键词：Credit risk,Multi-objective evolutionary algorithm,Classification rules

论文评审过程：Received 6 June 2018, Revised 25 December 2018, Accepted 30 January 2019, Available online 5 February 2019, Version of Record 23 February 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.01.078