Imbalance accuracy metric for model selection in multi-class imbalance classification problems

作者：

Highlights：

• The IAM is proposed as a metric for model selection in multi-class imbalance problems.

• The IAM is built up on top of the existing metrics and is simple to use.

• The IAM shows how well a classifier does not classify an instance in incorrect classes.

• The IAM is to eliminate the need for multiple metric computation in model selection.

摘要

•The IAM is proposed as a metric for model selection in multi-class imbalance problems.•The IAM is built up on top of the existing metrics and is simple to use.•The IAM shows how well a classifier does not classify an instance in incorrect classes.•The IAM is to eliminate the need for multiple metric computation in model selection.

论文关键词：Machine learning,Classification accuracy,Multi-class problems,Imbalance datasets,Knowledge discovery

论文评审过程：Received 28 May 2020, Revised 17 July 2020, Accepted 25 September 2020, Available online 28 September 2020, Version of Record 3 October 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106490