Parameter Optimization of Support Vector Machine Based on Combined Algorithm of QPSO and SA

Author

Shi Yan ; Li Xiao-min ; Qi Xiao-hui

Author_Institution

Dept. of Opt. & Electrics Eng., Ordnance Eng. Coll., Shijiazhuang, China

fYear

2010

fDate

17-19 Sept. 2010

Firstpage

483

Lastpage

486

Abstract

Support Vector Machine (SVM) is the focus of failure diagnose field. There is not a definite theory to guide the choice of its parameters. In this paper, the analysis and research is done to parameter optimization of SVM. The combined algorithm based on Quantum-behavior Particle Swarm Optimization (QPSO) and Simulated Annealing (SA) is present to optimize the parameters of SVM in order to improve the classification performance of SVM. The comparison of optimization result is done to other algorithms, it testifies that optimization effect of combined algorithm is better.

Keywords

particle swarm optimisation; pattern classification; quantum computing; simulated annealing; support vector machines; combined algorithm; failure diagnosis; parameter optimization; pattern classification; quantum behavior particle swarm optimization; simulated annealing; support vector machine; Accuracy; Algorithm design and analysis; Classification algorithms; Particle swarm optimization; Simulated annealing; Support vector machines; Parameter Optimization; Quantum-behavior Particle Swarm Optimization (QPSO); Simulated Annealing (SA); Support Vector Machine (SVM);

fLanguage

English

Publisher

ieee

Conference_Titel

Pervasive Computing Signal Processing and Applications (PCSPA), 2010 First International Conference on

Conference_Location

Harbin

Print_ISBN

978-1-4244-8043-2

Electronic_ISBN

978-0-7695-4180-8

Type

conf

DOI

10.1109/PCSPA.2010.122

Filename

5635513