Unstructured borderline self-organizing map: Learning highly imbalanced, high-dimensional datasets for fault detection

作者：

Highlights：

• A model is proposed to learn imbalanced, high-dimensional data for fault detection.

• Borderline areas are highlighted when learning the distribution of normal samples.

• The distributional change through learning reveals the importance of features.

• The superiority of the proposed model is verified via empirical case studies.

摘要

•A model is proposed to learn imbalanced, high-dimensional data for fault detection.•Borderline areas are highlighted when learning the distribution of normal samples.•The distributional change through learning reveals the importance of features.•The superiority of the proposed model is verified via empirical case studies.

论文关键词：Borderline,Fault detection,Feature selection,Industrial processes,Resampling,Self-organizing map

论文评审过程：Received 30 December 2020, Revised 16 July 2021, Accepted 1 October 2021, Available online 9 October 2021, Version of Record 14 October 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116028