NI-MWMOTE: An improving noise-immunity majority weighted minority oversampling technique for imbalanced classification problems
作者：
Highlights：
• New noise-immunity oversampling method for imbalanced classification problems.
• It adaptively processes noise based on probability and misclassification error.
• It simplifies the minority class clustering process using unsupervised clustering.
• It adaptively determines the sub-cluster size being sampled using misclassification error.
• Generated synthetic instances using MWMOTE improved subsequent classification.
摘要
•New noise-immunity oversampling method for imbalanced classification problems.•It adaptively processes noise based on probability and misclassification error.•It simplifies the minority class clustering process using unsupervised clustering.•It adaptively determines the sub-cluster size being sampled using misclassification error.•Generated synthetic instances using MWMOTE improved subsequent classification.
论文关键词：Imbalanced classification,Noise-immunity,MWMOTE,Clustering,Oversampling
论文评审过程：Received 4 September 2019, Revised 30 April 2020, Accepted 1 May 2020, Available online 4 May 2020, Version of Record 24 May 2020.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.113504
