F-Measure Maximization in Multi-Label Classification with Conditionally Independent Label Subsets

Maxime Gasse 1 Alex Aussem 1
1 DM2L - Data Mining and Machine Learning
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
Abstract : We discuss a method to improve the exact F-measure max-imization algorithm called GFM, proposed in [2] for multi-label classification , assuming the label set can be partitioned into conditionally independent subsets given the input features. If the labels were all independent , the estimation of only m parameters (m denoting the number of labels) would suffice to derive Bayes-optimal predictions in O(m^2) operations [10]. In the general case, m^2 + 1 parameters are required by GFM, to solve the problem in O(m^3) operations. In this work, we show that the number of parameters can be reduced further to m^2 /n, in the best case, assuming the label set can be partitioned into n conditionally independent subsets. As this label partition needs to be estimated from the data beforehand, we use first the procedure proposed in [4] that finds such partition and then infer the required parameters locally in each label subset. The latter are aggregated and serve as input to GFM to form the Bayes-optimal prediction. We show on a synthetic experiment that the reduction in the number of parameters brings about significant benefits in terms of performance.
Type de document :
Communication dans un congrès
Paolo Frasconi; Niels Landwehr; Giuseppe Manco; Jilles Vreeken. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2016, Riva del garda, Italy. Springer, Lecture Notes in Computer Science, 9851 (Part I), pp.619 - 631, 2016, Machine Learning and Knowledge Discovery in Databases. 〈http://link.springer.com/chapter/10.1007/978-3-319-46128-1_39〉. 〈10.1007/978-3-319-46128-1_39〉
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01425528
Contributeur : Maxime Gasse <>
Soumis le : mardi 3 janvier 2017 - 15:56:36
Dernière modification le : lundi 10 décembre 2018 - 17:45:10
Document(s) archivé(s) le : mardi 4 avril 2017 - 14:39:59

Fichier

ECML_2016 (1).pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Maxime Gasse, Alex Aussem. F-Measure Maximization in Multi-Label Classification with Conditionally Independent Label Subsets. Paolo Frasconi; Niels Landwehr; Giuseppe Manco; Jilles Vreeken. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2016, Riva del garda, Italy. Springer, Lecture Notes in Computer Science, 9851 (Part I), pp.619 - 631, 2016, Machine Learning and Knowledge Discovery in Databases. 〈http://link.springer.com/chapter/10.1007/978-3-319-46128-1_39〉. 〈10.1007/978-3-319-46128-1_39〉. 〈hal-01425528〉

Partager

Métriques

Consultations de la notice

352

Téléchargements de fichiers

237