A New Discernibility Metric and Its Application on Pattern Classification and Feature Evaluation

Abstract : A novel evaluation metric is introduced, based on the Discernibility concept. This metric, the Distance-based Index of Discernibility (DID) aims to provide an accurate and fast mapping of the classification performance of a feature or a dataset. DID has been successfully implemented in a program which has been applied to a number of datasets, a few artificial features and a typical benchmark dataset. The results appear to be quite promising, verifying the initial hypothesis.
Type de document :
Communication dans un congrès
Lazaros Iliadis; Ilias Maglogiannis; Harris Papadopoulos. 12th Engineering Applications of Neural Networks (EANN 2011) and 7th Artificial Intelligence Applications and Innovations (AIAI), Sep 2011, Corfu, Greece. Springer, IFIP Advances in Information and Communication Technology, AICT-364 (Part II), pp.27-35, 2011, Artificial Intelligence Applications and Innovations. 〈10.1007/978-3-642-23960-1_4〉
Liste complète des métadonnées

Littérature citée [11 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01571488
Contributeur : Hal Ifip <>
Soumis le : mercredi 2 août 2017 - 16:22:29
Dernière modification le : vendredi 1 décembre 2017 - 01:16:25

Fichier

978-3-642-23960-1_4_Chapter.pd...
Fichiers produits par l'(les) auteur(s)

Licence


Distributed under a Creative Commons Paternité 4.0 International License

Identifiants

Citation

Zacharias Voulgaris. A New Discernibility Metric and Its Application on Pattern Classification and Feature Evaluation. Lazaros Iliadis; Ilias Maglogiannis; Harris Papadopoulos. 12th Engineering Applications of Neural Networks (EANN 2011) and 7th Artificial Intelligence Applications and Innovations (AIAI), Sep 2011, Corfu, Greece. Springer, IFIP Advances in Information and Communication Technology, AICT-364 (Part II), pp.27-35, 2011, Artificial Intelligence Applications and Innovations. 〈10.1007/978-3-642-23960-1_4〉. 〈hal-01571488〉

Partager

Métriques

Consultations de la notice

103

Téléchargements de fichiers

5