Towards more efficient multiclass AUC computations

Publikation: Beitrag in Buch/Bericht/TagungsbandKonferenzbeitragBegutachtung

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.

OriginalspracheEnglisch
Titel31st European Modeling and Simulation Symposium, EMSS 2019
Redakteure/-innenMichael Affenzeller, Agostino G. Bruzzone, Francesco Longo, Guilherme Pereira
Herausgeber (Verlag)DIME UNIVERSITY OF GENOA
Seiten327-331
Seitenumfang5
ISBN (elektronisch)9788885741263
PublikationsstatusVeröffentlicht - 2019
Veranstaltung31st European Modeling and Simulation Symposium, EMSS 2019 - Lisbon, Portugal
Dauer: 18 Sep 201920 Sep 2019

Publikationsreihe

Name31st European Modeling and Simulation Symposium, EMSS 2019

Konferenz

Konferenz31st European Modeling and Simulation Symposium, EMSS 2019
Land/GebietPortugal
OrtLisbon
Zeitraum18.09.201920.09.2019

Fingerprint

Untersuchen Sie die Forschungsthemen von „Towards more efficient multiclass AUC computations“. Zusammen bilden sie einen einzigartigen Fingerprint.

Zitieren