Authors: E.A. Zanaty, Sultan Hamadi Aljahdali, R.J. Cripps
Addresses: College of Computer Science, Taif University, Hawya, P.O. Box 888, Taif, Saudi Arabia. ' College of Computer Science, Taif University, Hawya, P.O. Box 888, Taif, Saudi Arabia. ' Geometric Modeling Group, School Mechanical Engineering, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
Abstract: In this paper, a new kernel function is introduced that improves the classification accuracy of support vector machines (SVMs) for both linear and non-linear data sets. The proposed kernel function, called Gauss radial basis polynomial function (RBPF) combine both Gauss radial basis function (RBF) and polynomial (POLY) kernels. It is shown that the proposed kernel converges faster than the RBF and POLY kernels. The accuracy of the proposed algorithm is compared to algorithms based on both Gaussian and polynomial kernels by application to a variety of non-separable data sets with several attributes. It is shown that the proposed kernel gives good classification accuracy in nearly all the data sets, especially those of high dimensions.
Keywords: kernel functions; support vector machines; SVMs; classification accuracy; Gauss radial basis function; kernels; polynomial kernels.
International Journal of Rapid Manufacturing, 2009 Vol.1 No.2, pp.114 - 127
Available online: 03 Dec 2009 *Full-text access for editors Access for subscribers Purchase this article Comment on this article