Multiclass Classification Performance Curve

doi:10.1109/ACCESS.2022.3186444

Multiclass Classification Performance Curve

2022-01-01

8

OA

AI

J

Jesús S. Aguilar–Ruiz *

M

Marcin Michalak

DOI：10.1109/ACCESS.2022.3186444

Original

Original request for help

Save

Abstract

En 中文

Quality of predictive models is a critical factor. Many evaluation measures have been proposed for binary and multi-class datasets. However, less attention has been paid to graphical representation of the classification performance, where the ROC curve is extensively used for binary datasets but there is no standard method accepted by the scientific community for multi-class datasets. In this work, a multi-class classification performance (MCP) curve based on the Hellinger distance between true and prediction probabilities of the classifier is introduced. The MCP curve shows the classification performance, contributes to highlight the low or high confidence on correct predictions, and quantifies the quality by means of the area under the curve.

Keywords:

Predictive models

Biomedical measurement

Sensitivity

Probability distribution

Medical diagnostic imaging

Licenses

Correlation coefficient

Classification

machine learning

multi-class data

performance curve

predictive models

ROC curve

Multiclass Classification Performance Curve

Abstract

Journal

Organization