Model parameter selection of support vector machines

Author

Zhao, Mingyuan ; Tang, Ke ; Zhou, Mingtian ; Zhang, Fengli ; Zeng, Ling

Author_Institution

Sch. of Comput. Sci. & Technol., Univ. of Electron. Sci. & Technol. of China, Chengdu

fYear

2008

fDate

21-24 Sept. 2008

Firstpage

1095

Lastpage

1099

Abstract

In order to optimize classification performance of support vector machines, analyzing character of model parameters on support vector machines with Gaussian kernel, using data of Ionosphere Database in UCI repository of machine learning database and electroencephalogram (EEG) experiment data to make and analyze the area search table, a new parametric distribution model is proposed. In order to search optimal points in model parameters of support vector machines, a new genetic algorithm based on parametric distribution model is proposed to improve classification performance of support vector machines remarkably.

Keywords

genetic algorithms; learning (artificial intelligence); pattern classification; support vector machines; Gaussian kernel; classification performance; electroencephalogram experiment data; genetic algorithm; machine learning database; model parameter selection; parametric distribution model; support vector machines; Brain modeling; Databases; Electroencephalography; Genetic algorithms; Ionosphere; Kernel; Machine learning; Performance analysis; Support vector machine classification; Support vector machines; area search table; genetic algorithm; parameters selection; parametric distribution model; support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Cybernetics and Intelligent Systems, 2008 IEEE Conference on

Conference_Location

Chengdu

Print_ISBN

978-1-4244-1673-8

Electronic_ISBN

978-1-4244-1674-5

Type

conf

DOI

10.1109/ICCIS.2008.4670757

Filename

4670757