Simulated annealing least squares twin support vector machine (SA-LSTSVM) for pattern classification

Sartakhti, JS; Afrabandpey, H; Saraee, MH

Abstract

Least squares twin support vector machine
(LSTSVM) is a relatively new version of support vector
machine (SVM) based on non-parallel twin hyperplanes.
Although, LSTSVM is an extremely efficient and fast algorithm for binary classification, its parameters depend on the nature of the problem. Problem dependent parameters
make the process of tuning the algorithm with best values
for parameters very difficult, which affects the accuracy of the algorithm. Simulated annealing (SA) is a random search technique proposed to find the global minimum of a cost function. It works by emulating the process where a metal slowly cooled so that its structure finally “freezes”. This freezing point happens at a minimum energy configuration. The goal of this paper is to improve the accuracy of the LSTSVMalgorithmby hybridizing it with simulated anneaing. Our research to date suggests that this improvement on the LSTSVM is made for the first time in this paper. Experimental results on several benchmark datasets demonstrate that the accuracy of the proposed algorithm is very promising when compared to other classification methods in the literature. In addition, computational time analysis of the algorithm showed the practicality of the proposed algorithm where the computational time of the algorithm falls between
LSTSVM and SVM.

Citation

Sartakhti, J., Afrabandpey, H., & Saraee, M. (2017). Simulated annealing least squares twin support vector machine (SA-LSTSVM) for pattern classification. Soft Computing, 21(15), 4361-4373. https://doi.org/10.1007/s00500-016-2067-4

• Journal Article Type: Article
• Online Publication Date: Feb 25, 2016
• Publication Date: Aug 1, 2017
• Deposit Date: Feb 9, 2016
• Publicly Available Date: Feb 25, 2017
• Journal: Soft Computing - A Fusion of Foundations, Methodologies and Applications
• Print ISSN: 1432-7643
• Electronic ISSN: 1433-7479
• Publisher: Springer Verlag
• Volume: 21
• Issue: 15
• Pages: 4361-4373
• DOI: https://doi.org/10.1007/s00500-016-2067-4
• Publisher URL: http://dx.doi.org/10.1007/s00500-016-2067-4
• Related Public URLs: http://link.springer.com/journal/500