Genetic Algorithm based Comparison of Different SVM

The SVM has recently been introduced as a new learning technique for solving variety of real world applications based on learning theory. The classical RBF network has similar structure as SVM with Gaussian kernel. Similarly, the FNN also possess an identical structure with SVM. The support vector machine includes polynomial learning machine, radial-basis function network, Gaussian radial-basis function network, and two layer perceptron as special cases. Genetic algorithm has been increasingly applied to various search and optimization problems in the recent past but in spite of its broad applicability, ease of use and global perspective, it has not yet been used in comparison and optimization of different support vector machines. In this paper attempt has been made to compare and optimize the rate of convergence of different SVMs by using the concepts of GAs and important results are worked out.