A minority oversampling approach for fault detection with heterogeneous imbalanced data

作者：

Highlights：

• Feature heterogeneity problem in imbalanced data is considered.

• Two new minority oversampling methods are proposed.

• Various public imbalanced datasets are considered in the experiment.

• A real case study on fault detection is considered.

• Results show the effectiveness of the proposed methods.

摘要

•Feature heterogeneity problem in imbalanced data is considered.•Two new minority oversampling methods are proposed.•Various public imbalanced datasets are considered in the experiment.•A real case study on fault detection is considered.•Results show the effectiveness of the proposed methods.

论文关键词：Fault detection,Imbalanced data,Feature heterogeneity,Minority oversampling

论文评审过程：Received 30 August 2020, Revised 30 September 2020, Accepted 24 June 2021, Available online 1 July 2021, Version of Record 6 July 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115492