Improving SVM Training Sample Selection Using Multi-Objective Evolutionary Algorithm and LSH

R. Pighetti Denis Pallez 1 Frédéric Precioso 2
1 I3S - Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis
2 Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe KEIA
Laboratoire I3S - SPARKS - Scalable and Pervasive softwARe and Knowledge Systems
Abstract : In this paper, we propose a new framework hybridizing a Support Vector Machine (SVM), a Multi-Objective Genetic Algorithm (MOGA) and a Locality Sensitive Hashing (LSH). The goal is to tackle fine-grained classification challenges which means classifying many classes with high similarities between classes and poor similarities inside one class. SVM is used for its ability of learning multi-class problem from very few training data. MOGA is used for optimizing training samples used by SVM so as to improve its learning rate. As data define a discrete set of instances and not a continuous solution space, LSH is used for mapping "optimal solutions" obtained by MOGA onto the closest real instances contained in the dataset. We evaluate our method for content-based image classification on the standard image database Caltech256 (i.e. 30000 images distributed in 256 classes). Experiments shows that our method outperforms reference approaches.
Keywords : Training data Training Buildings support vector machine multiobjective genetic algorithm Context Evolutionary computation genetic algorithms image classification support vector machines LSH MOGA SVM training sample selection content-based image classification fine-grained classification locality sensitive hashing multiobjective evolutionary algorithm
Type de document :
Communication dans un congrès
Computational Intelligence, 2015 IEEE Symposium Series on, 2015, Cape Town, South Africa. pp.1383-1390, 2015, Computational Intelligence and Data Mining. 〈10.1109/SSCI.2015.197〉
Liste complète des métadonnées
https://hal.univ-cotedazur.fr/hal-01322765
Contributeur : Denis Pallez <>
Soumis le : vendredi 27 mai 2016 - 16:17:16
Dernière modification le : lundi 5 novembre 2018 - 15:52:10

Identifiants

Collections

UNICE | I3S | UNIV-COTEDAZUR

Citation

R. Pighetti, Denis Pallez, Frédéric Precioso. Improving SVM Training Sample Selection Using Multi-Objective Evolutionary Algorithm and LSH. Computational Intelligence, 2015 IEEE Symposium Series on, 2015, Cape Town, South Africa. pp.1383-1390, 2015, Computational Intelligence and Data Mining. 〈10.1109/SSCI.2015.197〉. 〈hal-01322765〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

160

Contact

Université Nice Sophia Antipolis (UNS) Observatoire de la Côte d’Azur (OCA) Centre national de la recherche scientifique (CNRS) Institut national de la recherche en informatique et en automatique (Inria) SKEMA Business School EDHEC Business School  Centre Hospitalier Universitaire de Nice (CHU Nice) Centre National de Création Musicale (CIRM)  Ecole Nationale Supérieure d’Art Villa Arson Ecole Supérieure de Réalisation Audiovisuelle (ESRA) The Sustainable Design School (SDS) Ecole Supérieure de Danse de Cannes Rosella Hightower Conservatoire National à Rayonnement Régional de Nice (CNRR)