An Implementation of Support Vector Machines on Iris Dataset

Abstract

Support Vector Machines (SVM) is a set of related supervised learning method used for classification. SVM is used to construct a hyperplane as the decision surface in such a way that the margin of separation between positive and negative examples is maximized. By default, this hyperplane is linear. To improve the classification performance, it is desirable to use a non-linear hyperplane. In order to construct a non-linear hyperplane using SVM, we use kernel functions. This paper presents a comparison of using several kernel functions in the SVM algorithm for Iris dataset classification.