Differential evolution-based parameters optimisation and feature selection for support vector machine. (2016)

• Record Type: Journal Article
• Authors: Li, Jun; Ding, Lixin; Li, Bo
• Is Part Of: International journal of computational science and engineering. Volume 13:Number 4(2016)
• Page Start: 355
• Page End: 363
• Publication Date: 2016
• Subjects: high-dimensional classification -- support vector machines -- SVM -- differential evolution -- parameter optimisation -- feature selection; Computer science -- Mathematics -- Periodicals; Computer simulation -- Mathematical aspects -- Periodicals; Computational intelligence -- Periodicals; 004.015105
• Languages: English
• ISSNs: 1742-7185

Abstract

This paper addresses the problem of SVM classification optimisation. For this purpose, the authors propose an SVM classification system based on differential evolution (DE) to improve the generalisation performance of the SVM classifier. In the classification system, a method of simultaneous parameters optimisation and feature selection for support vector machine is put forward. The experiments are conducted on the basis of benchmark dataset. The obtained results clearly confirm the superiority of the DE-SVM-FS approach compared to default SVM classifier and DE-SVM algorithm; this suggests that further substantial improvements in terms of classification accuracy can be achieved by the proposed DE-SVM-FS classification system.