Radial-Based Undersampling for imbalanced data classification

作者：

Highlights：

• The concept of mutual class potential is extended to the undersampling procedure.

• Radial-Based Undersampling offers significantly lower computational complexity.

• Method achieves significantly better results when combined with selected classifiers.

• Areas of applicability of the algorithm are identified.

摘要

•The concept of mutual class potential is extended to the undersampling procedure.•Radial-Based Undersampling offers significantly lower computational complexity.•Method achieves significantly better results when combined with selected classifiers.•Areas of applicability of the algorithm are identified.

论文关键词：Machine learning,Classification,Imbalanced data,Undersampling,Radial basis functions

论文评审过程：Received 10 June 2019, Revised 15 November 2019, Accepted 5 February 2020, Available online 5 February 2020, Version of Record 12 February 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107262