A comparative study and performance analysis of classification techniques: Support vector machine, neural networks and decision trees

Abstract

A support vector machine (SVM) is a classification technique in the field of data mining, used for the classification of both linear as well as non-linear data. It learns the decision surface from two different classes of input samples and then performs analysis of new input samples. A neural network is able to learn without the explicit description of the problem or the need of a programmer. Another type of classification technique is the decision tree. In this paper, we are doing a comparative study of the above mentioned classification techniques by analyzing their performance on data sets. We will be comparing the inputs and the observed outputs.