Hierarchical Behavior Knowledge Space

Hubert Cecotti 1 Abdel Belaïd 2
2 READ - READ
LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : In this paper we present a new method for fusing classiers output for problems with a number of classes M > 2. We extend the well-known Behavior Knowledge Space method with a hierarchical ap- proach of the dierent cells. We propose to add the ranking information of the classiers output for the combination. Each cell can be divided into new sub-spaces in order to solve ambiguities. We show that this method allows a better control of the rejection, without using new classiers for the empty cells. This method has been applied on a set of classi- ers created by bagging. It has been successfully tested on handwritten character recognition allowing better-detailed results. The technique has been compared with other classical combination methods.
Type de document :
Communication dans un congrès
7th International Workshop on Multiple Classier Systems - MCS 2007, May 2007, Prague, Czech Republic. Springer, 4472, pp.421-430, 2007, Lecture Notes in Computer Science. 〈10.1007/978-3-540-72523-7_42〉
Liste complète des métadonnées

Littérature citée [11 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/inria-00579636
Contributeur : Abdel Belaid <>
Soumis le : jeudi 24 mars 2011 - 14:40:39
Dernière modification le : jeudi 11 janvier 2018 - 06:19:59
Document(s) archivé(s) le : samedi 25 juin 2011 - 02:35:07

Fichier

cecotti-mcs07.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Hubert Cecotti, Abdel Belaïd. Hierarchical Behavior Knowledge Space. 7th International Workshop on Multiple Classier Systems - MCS 2007, May 2007, Prague, Czech Republic. Springer, 4472, pp.421-430, 2007, Lecture Notes in Computer Science. 〈10.1007/978-3-540-72523-7_42〉. 〈inria-00579636〉

Partager

Métriques

Consultations de la notice

136

Téléchargements de fichiers

91