Model Selection in SVMs Using Differential Evolution

Jingjing Zhang; Qun Niu; Kang Li; George W. Irwin

doi:10.3182/20110828-6-IT-1002.00584

Model Selection in SVMs Using Differential Evolution

Jingjing Zhang, Qun Niu, Kang Li, George W. Irwin

Research output: Contribution to conference › Paper

6 Citations (Scopus)

Abstract

To improve the performance of classification using Support Vector Machines (SVMs) while reducing the model selection time, this paper introduces Differential Evolution, a heuristic method for model selection in two-class SVMs with a RBF kernel. The model selection method and related tuning algorithm are both presented. Experimental results from application to a selection of benchmark datasets for SVMs show that this method can produce an optimized classification in less time and with higher accuracy than a classical grid search. Comparison with a Particle Swarm Optimization (PSO) based alternative is also included.

Original language	English
Pages	14717-14722
Number of pages	6
DOIs	https://doi.org/10.3182/20110828-6-IT-1002.00584
Publication status	Published - 2011
Event	18th World Congress of the International Federation of Automatic Control 2011 - Milan, Italy Duration: 28 Aug 2011 → 02 Sept 2011

Conference

Conference	18th World Congress of the International Federation of Automatic Control 2011
Abbreviated title	IFAC 2011
Country/Territory	Italy
City	Milan
Period	28/08/2011 → 02/09/2011

Bibliographical note

ISSN: 978-3-902661-93-7

Keywords

Fuzzy and neural systems relevant to control and identification
Evolutionary algorithms in control and identification

Access to Document

10.3182/20110828-6-IT-1002.00584

Cite this

@conference{6b32341cfc054ebdaf46ca86f2b26a16,

title = "Model Selection in SVMs Using Differential Evolution",

abstract = "To improve the performance of classification using Support Vector Machines (SVMs) while reducing the model selection time, this paper introduces Differential Evolution, a heuristic method for model selection in two-class SVMs with a RBF kernel. The model selection method and related tuning algorithm are both presented. Experimental results from application to a selection of benchmark datasets for SVMs show that this method can produce an optimized classification in less time and with higher accuracy than a classical grid search. Comparison with a Particle Swarm Optimization (PSO) based alternative is also included.",

keywords = "Fuzzy and neural systems relevant to control and identification, Evolutionary algorithms in control and identification",

author = "Jingjing Zhang and Qun Niu and Kang Li and Irwin, \{George W.\}",

note = "ISSN: 978-3-902661-93-7; 18th World Congress of the International Federation of Automatic Control 2011, IFAC 2011 ; Conference date: 28-08-2011 Through 02-09-2011",

year = "2011",

doi = "10.3182/20110828-6-IT-1002.00584",

language = "English",

pages = "14717--14722",

}

TY - CONF

T1 - Model Selection in SVMs Using Differential Evolution

AU - Zhang, Jingjing

AU - Niu, Qun

AU - Li, Kang

AU - Irwin, George W.

N1 - ISSN: 978-3-902661-93-7

PY - 2011

Y1 - 2011

N2 - To improve the performance of classification using Support Vector Machines (SVMs) while reducing the model selection time, this paper introduces Differential Evolution, a heuristic method for model selection in two-class SVMs with a RBF kernel. The model selection method and related tuning algorithm are both presented. Experimental results from application to a selection of benchmark datasets for SVMs show that this method can produce an optimized classification in less time and with higher accuracy than a classical grid search. Comparison with a Particle Swarm Optimization (PSO) based alternative is also included.

AB - To improve the performance of classification using Support Vector Machines (SVMs) while reducing the model selection time, this paper introduces Differential Evolution, a heuristic method for model selection in two-class SVMs with a RBF kernel. The model selection method and related tuning algorithm are both presented. Experimental results from application to a selection of benchmark datasets for SVMs show that this method can produce an optimized classification in less time and with higher accuracy than a classical grid search. Comparison with a Particle Swarm Optimization (PSO) based alternative is also included.

KW - Fuzzy and neural systems relevant to control and identification

KW - Evolutionary algorithms in control and identification

U2 - 10.3182/20110828-6-IT-1002.00584

DO - 10.3182/20110828-6-IT-1002.00584

M3 - Paper

SP - 14717

EP - 14722

T2 - 18th World Congress of the International Federation of Automatic Control 2011

Y2 - 28 August 2011 through 2 September 2011

ER -

Model Selection in SVMs Using Differential Evolution

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this