Bankruptcy prediction on the base of the unbalanced data using multi-objective selection of classifiers

作者：

Highlights：

• The new method of classifier selection for imbalanced data is proposed.

• Selection rule uses the dominance in the FPR/FNR space and crowding distance.

• Tests on two real datasets confirm that new method improves the prediction.

摘要

•The new method of classifier selection for imbalanced data is proposed.•Selection rule uses the dominance in the FPR/FNR space and crowding distance.•Tests on two real datasets confirm that new method improves the prediction.

论文关键词：Bankruptcy prediction,Financial failure prediction,Unbalanced data classification,False positive rate,False negative rate

论文评审过程：Received 1 March 2020, Revised 24 March 2021, Accepted 3 July 2021, Available online 17 July 2021, Version of Record 20 July 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115559