Hierarchical Label Partitioning for Large Scale Classification

Raphael Puget 1 Nicolas Baskiotis 1
1 MLIA - Machine Learning and Information Access
LIP6 - Laboratoire d'Informatique de Paris 6
Abstract : Extreme classification task where the number of classes is very large has received important focus over the last decade. Usual efficient multi-class classification approaches have not been designed to deal with such large number of classes. A particular issue in the context of large scale problems concerns the computational classification complexity : best multi-class approaches have generally a linear complexity with respect to the number of classes which does not allow these approaches to scale up. Recent works have put their focus on using hierarchical classification process in order to speed-up the classification of new instances. A priori information on labels is not always available nor useful to build hierarchical models. Finding a suitable hierarchical organization of the labels is thus a crucial issue as the accuracy of the model depends highly on the label assignment through the label tree. We propose in this work a new algorithm to build iteratively a hierarchical label structure by proposing a partitioning algorithm which optimizes simultaneously the structure in terms of classification complexity and the label partitioning problem in order to achieve high classification performances. Beginning from a flat tree structure, our algorithm selects iteratively a node to expand by adding a new level of nodes between the considered node and its children. This operation increases the speed-up of the classification process. Once the node is selected, best partitioning of the classes has to be computed. We propose to consider a measure based on the maximization of the expected loss of the sub-levels in order to minimize the global error of the structure. This choice enforces hardly separable classes to be group together in same partitions at the first levels of the tree structure and it delays errors at a deep level of the structure where there is no incidence on the accuracy of other classes.
Type de document :
Communication dans un congrès
IEEE International Conference on Data Science and Advanced Analytics, DSAA'2015, Oct 2015, Paris, France. 2015
Liste complète des métadonnées

Littérature citée [36 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01207430
Contributeur : Raphael Puget <>
Soumis le : jeudi 1 octobre 2015 - 15:54:07
Dernière modification le : jeudi 22 novembre 2018 - 14:23:21
Document(s) archivé(s) le : samedi 2 janvier 2016 - 10:20:56

Fichier

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

Licence


Copyright (Tous droits réservés)

Identifiants

  • HAL Id : hal-01207430, version 1

Collections

Citation

Raphael Puget, Nicolas Baskiotis. Hierarchical Label Partitioning for Large Scale Classification. IEEE International Conference on Data Science and Advanced Analytics, DSAA'2015, Oct 2015, Paris, France. 2015. 〈hal-01207430〉

Partager

Métriques

Consultations de la notice

152

Téléchargements de fichiers

384