Accuracy Measures for the Comparison of Classifiers

Abstract : The selection of the best classification algorithm for a given dataset is a very widespread problem. It is also a complex one, in the sense it requires to make several important methodological choices. Among them, in this work we focus on the measure used to assess the classification performance and rank the algorithms. We present the most popular measures and discuss their properties. Despite the numerous measures proposed over the years, many of them turn out to be equivalent in this specific case, to have interpretation problems, or to be unsuitable for our purpose. Consequently, classic overall success rate or marginal rates should be preferred for this specific task.
Type de document :
Communication dans un congrès
Al-Dahoud Ali. The 5th International Conference on Information Technology, May 2011, amman, Jordan. Al-Zaytoonah University of Jordan, pp.1,5, 2011
Liste complète des métadonnées

Littérature citée [28 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-00611319
Contributeur : Hocine Cherifi <>
Soumis le : jeudi 28 juillet 2011 - 11:24:59
Dernière modification le : lundi 16 juillet 2012 - 10:49:34
Document(s) archivé(s) le : lundi 7 novembre 2011 - 12:22:00

Fichier

17_ICIT11_VL.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00611319, version 1
  • ARXIV : 1207.3790

Collections

Citation

Vincent Labatut, Hocine Cherifi. Accuracy Measures for the Comparison of Classifiers. Al-Dahoud Ali. The 5th International Conference on Information Technology, May 2011, amman, Jordan. Al-Zaytoonah University of Jordan, pp.1,5, 2011. 〈hal-00611319〉

Partager

Métriques

Consultations de la notice

220

Téléchargements de fichiers

4755