Evolutionary design of multiclass support vector machines

Issue title: VIII Brazilian Symposium on Neural Networks

Article type: Research Article

Authors: Lorena, Ana C. | Carvalho, André C. P. L. F.

Affiliations: Departamento de Ciências de Computação, Instituto de Ciências Matemáticas e de Computação, Universidade de Sáo Paulo, campus de São Carlos, Caixa Postal 668, CEP 13560–970, São Carlos, SP, Brazil

Note: [] Corresponding author. Tel.: +55 16 3373 9646; Fax: +55 16 3371 2238; E-mail: aclorena@icmc.usp.br

Abstract: Support Vector Machines constitute a Machine Learning technique originally designed for the solution of 2-class problems. For multiclass applications, several strategies divide the original problem into a set of binary subtasks, whose results are combined. This work introduces the use of Genetic Algorithms to determine binary decompositions of multiclass problems. Experimental results on benchmark and Bioinformatics multiclass datasets indicate the potential of the proposed approach, which is able to produce good multiclass solutions with the use of simple decompositions.

Keywords: Support vector machines, multiclass classification

Journal: Journal of Intelligent & Fuzzy Systems, vol. 18, no. 5, pp. 445-454, 2007