Nonlinear Classifier Design Research Based on SVM and Genetic Algorithm

Wenjuan Zeng (Hunan International Economics University)
Haibo Gao (Hunan International Economics University)

Indonesian Journal of Electrical Engineering and Computer Science
Vol 13, No 1: January 2015

This paper presents a support vector machine (SVM) model structure, the genetic algorithm parameters of the model portfolio optimization model, and used for non-linear pattern recognition, the method is not only effective for linear problems, nonlinear problems apply effective; the law simple and easy, better than the multi-segment linear classifier design methods and BP network algorithm returns the error. Examples show the efficiency of its recognition of 100%. DOI:  http://dx.doi.org/10.11591/telkomnika.v13i1.6692