IA-SUWO: An Improving Adaptive semi-unsupervised weighted oversampling for imbalanced classification problems

作者：

Highlights：

• IA-SUWO is proposed to address the imbalanced classification problems.

• IA-SUWO uses a new system based on the IMWMO method and the -method to assign weights to minority instances.

• IA-SUWO uses a new method based on k* information nearest neighbors to generate minority instances.

• Results show that IA-SUWO is superior to the 12 existing algorithms.

摘要

•IA-SUWO is proposed to address the imbalanced classification problems.•IA-SUWO uses a new system based on the IMWMO method and the -method to assign weights to minority instances.•IA-SUWO uses a new method based on k* information nearest neighbors to generate minority instances.•Results show that IA-SUWO is superior to the 12 existing algorithms.

论文关键词：Imbalanced classification,Least squares support numerical spectrum,Minority samples weights,Oversampling,k* information nearest neighbors

论文评审过程：Received 4 November 2019, Revised 9 March 2020, Accepted 5 June 2020, Available online 10 June 2020, Version of Record 20 June 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106116