Towards more efficient multiclass AUC computations

Research output: Chapter in Book/Report/Conference proceedingsConference contributionpeer-review

Abstract

The area under the receiver operating characteristics curve (AUC) can be used to assess the discriminatory power of a dichotomous classifier model. Extending this measure to more than two classes is not obvious, and a number of variants have been proposed in the literature. We investigate a heuristic approximation to a method that generalizes the notion of probabilities being correctly ordered, which is equivalent to AUC, to an arbitrary number of classes. While the exact method is computationally complex, we propose a much simpler heuristic that is linear in the number of classes for every combination of data points. Using one artificial and one real-world data set, we demonstrate empirically that this simple heuristic can provide good approximations to the exact method, with Pearson correlation coefficients between 0.85 and 0.998 across all data sets.

Original languageEnglish
Title of host publication31st European Modeling and Simulation Symposium, EMSS 2019
EditorsMichael Affenzeller, Agostino G. Bruzzone, Francesco Longo, Guilherme Pereira
PublisherDIME UNIVERSITY OF GENOA
Pages327-331
Number of pages5
ISBN (Electronic)9788885741263
Publication statusPublished - 2019
Event31st European Modeling and Simulation Symposium, EMSS 2019 - Lisbon, Portugal
Duration: 18 Sep 201920 Sep 2019

Publication series

Name31st European Modeling and Simulation Symposium, EMSS 2019

Conference

Conference31st European Modeling and Simulation Symposium, EMSS 2019
Country/TerritoryPortugal
CityLisbon
Period18.09.201920.09.2019

Keywords

  • Classifier performance assessment
  • Multiclass AUC
  • Multiclass ROC

Fingerprint

Dive into the research topics of 'Towards more efficient multiclass AUC computations'. Together they form a unique fingerprint.

Cite this