MLP and SVM networks - a comparative study

Abstract

The paper presents the comparative analysis of two most important neural networks: the multilayer perceptron (MLP) and Support Vector Machine (SVM). The most effective learning algorithms have been implemented in an uniform way using Matlab platform and undergone the comparison with respect to the complexity of the structure as well as the accuracy and the calculation time for the solution of different learning tasks, including classification, prediction and regression. The results of numerical experiments are given and discussed in the paper.