Generic performance measure for multiclass-classifiers

作者：

Highlights：

• Generic, compact and meaningful performance measure for arbitrary classifiers.

• Includes confidence indicators in interval [0; 1].

• Robust towards class imbalance, sensitive towards class separation.

• Demo implementation available.

摘要

•Generic, compact and meaningful performance measure for arbitrary classifiers.•Includes confidence indicators in interval [0; 1].•Robust towards class imbalance, sensitive towards class separation.•Demo implementation available.

论文关键词：Classification,Multiclass,Performance,Evaluation,Model selection,Accuracy

论文评审过程：Received 24 May 2016, Revised 2 March 2017, Accepted 4 March 2017, Available online 6 March 2017, Version of Record 22 April 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.03.008