A Survey on Hybrid Techniques Using SVM

Abstract

Support Vector Machines (SVM) with linear or nonlinear kernels has become one of the most promising learning algorithms for classification as well as for regression. All the multilayer perceptron (MLP),Radial Basic Function(RBF) and Learning Polynomials are also worked efficiently with SVM. SVM is basically derived from statistical Learning Theory and it is very powerful statistical tool. The basic principal for the SVM is structural risk minimization and closely related to regularization theory. SVM is a group of supervised learning techniques or methods, which is used to do for classification or regression. In this paper discussed the importance of Support Vector Machines in various areas. This paper discussing the efficiency of SVM with the combination of other classification techniques.