Adaptive Classifier Selection in Large-Scale Hierarchical Classification

Abstract : Going beyond the traditional text classification, involving a few tens of classes, there has been a surge of interest in automatic document categorization in large taxonomies where the number of classes range from hundreds of thousands to millions. Due to the complex nature of the learning problem posed in such scenarios, one needs to adapt the conventional classification schemes to suit this domain. This paper presents a novel approach for classifier selection in large hierarchies, which is based on exploiting training data heterogeneity across the hierarchy. We also present a meta-learning framework for further flexibility in classifier selection. The experimental results demonstrate the applicability of our approach, which achieves accuracy comparable to the state-of-the-art and is also significantly faster for prediction.
Type de document :
Communication dans un congrès
Tingwen Huang, Zhigang Zeng, Chuandong Li, Chi Sing Leung. ICONIP 2012 - International Conference on Neural Information Processing, Nov 2012, Doha, Qatar. Springer Berlin, 7665, pp.612-619, 2012, Lecture Notes in Computer Science (LNCS). <10.1007/978-3-642-34487-9_74>


https://hal.archives-ouvertes.fr/hal-00750771
Contributeur : Eric Gaussier <>
Soumis le : lundi 12 novembre 2012 - 15:17:03
Dernière modification le : mardi 28 octobre 2014 - 18:34:42
Document(s) archivé(s) le : mercredi 13 février 2013 - 03:44:38

Fichier

Partalas-Babbar-Gaussier-Ambla...
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Ioannis Partalas, Rohit Babbar, Éric Gaussier, Cécile Amblard. Adaptive Classifier Selection in Large-Scale Hierarchical Classification. Tingwen Huang, Zhigang Zeng, Chuandong Li, Chi Sing Leung. ICONIP 2012 - International Conference on Neural Information Processing, Nov 2012, Doha, Qatar. Springer Berlin, 7665, pp.612-619, 2012, Lecture Notes in Computer Science (LNCS). <10.1007/978-3-642-34487-9_74>. <hal-00750771>

Exporter

Partager

Métriques

Consultations de
la notice

189

Téléchargements du document

126