Training Multiclass Classifiers by Maximizing the Volume under the ROC Surface

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

2 Citations (Scopus)

Abstract

Receiver operating characteristic (ROC) curves are a plot of a ranking classifier's true-positive rate versus its false-positive rate, as one varies the threshold between positive and negative classifications across the continuum. The area under the ROC curve offer a measure of the discriminatory power of machine learning algorithms that is independent of class distribution, via its equivalence to Mann-Whitney U-statistics. This measure has recently been extended to cover problems of discriminating three and more classes. In this case, the area under the curve generalizes to the volume under the ROC surface. In this paper, we show how a multi-class classifier can be trained by directly maximizing the volume under the ROC surface. This is accomplished by first approximating the discrete U-statistic that is equivalent to the volume under the surface in a continuous manner, and then maximizing this approximation by gradient ascent.

Original languageEnglish
Title of host publicationComputer Aided Systems Theory - EUROCAST 2007 - 11th International Conference on Computer Aided Systems Theory, Revised Selected Papers
PublisherIUCTC Las Palmas de Gran Canaria
Pages878-885
Number of pages8
ISBN (Print)9783540758662
DOIs
Publication statusPublished - 2007
EventInternational Conference Computer Aided Systems Theory EUROCAST 2007 - Las Palmas, Spain
Duration: 12 Feb 200717 Feb 2007
http://www.ulpgc.es

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4739 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceInternational Conference Computer Aided Systems Theory EUROCAST 2007
Country/TerritorySpain
CityLas Palmas
Period12.02.200717.02.2007
Internet address

Keywords

  • Discrimination analysis
  • Machine learning
  • Multi-class ROC analysis

Fingerprint

Dive into the research topics of 'Training Multiclass Classifiers by Maximizing the Volume under the ROC Surface'. Together they form a unique fingerprint.

Cite this