Autor(es)

Pereira, Luis A. M. ; Papa, Joao Paulo ; de Souza, Andre N. ; Kuzle, I ; Capuder, T ; Pandzic, H

• Tipo de documento: Artigo Científico
• Fonte: 2013 IEEE EUROCON; v. N/A, p. 5-pg., 2013-01-01.

Resumo

Since the beginning, some pattern recognition techniques have faced the problem of high computational burden for dataset learning. Among the most widely used techniques, we may highlight Support Vector Machines (SVM), which have obtained very promising results for data classification. However, this classifier requires an expensive training phase, which is dominated by a parameter optimization that aims to make SVM less prone to errors over the training set. In this paper, we model the problem of finding such parameters as a metaheuristic-based optimization task, which is performed through Harmony Search (HS) and some of its variants. The experimental results have showen the robustness of HS-based approaches for such task in comparison against with an exhaustive (grid) search, and also a Particle Swarm Optimization-based implementation. (AU)

• Processo FAPESP: 12/14158-2 - Caracterização de Perdas Comerciais em Sistemas de Distribuição de Energia Utilizando Floresta de Caminhos Ótimos e Abordagens Evolucionistas
• Beneficiário: André Nunes de Souza
• Modalidade de apoio: Auxílio à Pesquisa - Regular
• Processo FAPESP: 09/16206-1 - Novas tendências em reconhecimento de padrões baseado em floresta de caminhos ótimos
• Beneficiário: João Paulo Papa
• Modalidade de apoio: Auxílio à Pesquisa - Jovens Pesquisadores
• Processo FAPESP: 11/14094-1 - Explorando Abordagens de Múltiplos Rótulos por Floresta de Caminhos Ótimos
• Beneficiário: Luis Augusto Martins Pereira
• Modalidade de apoio: Bolsas no Brasil - Mestrado