ROC representation for the discriminability of multi-classification markers

作者：

Highlights：

• A proper assessment of performance probabilities is shown to be the optimal ROC manifold.

• The necessary and sufficient conditions are addressed for the existence of the HUM.

• The scientific groundwork is extended to a more general multi-class ROC analysis.

摘要

Highlights•A proper assessment of performance probabilities is shown to be the optimal ROC manifold.•The necessary and sufficient conditions are addressed for the existence of the HUM.•The scientific groundwork is extended to a more general multi-class ROC analysis.

论文关键词：Discriminability,Hypervolume,Manifold,Optimal classification,Receiver operating characteristic,Utility

论文评审过程：Received 1 December 2014, Revised 9 April 2016, Accepted 26 June 2016, Available online 1 July 2016, Version of Record 16 July 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.06.024