Comparison of Single and Ensemble Classifiers of Support Vector Machine and Classification Tree

Abstract

An ensemble consists of a set of individually trained classifiers (such as Support Vector Machine and Classification Tree) whose predictions are combined by an algorithm. Ensemble methods is expected to improve the predictive performance of classifier. This research aims to assess and compare performance of single and ensemble classifiers of Support Vector Machine (SVM) and Classification Tree (CT) by using simulation data. The simulation data is based on three data structures which are linearly separable, linearly nonseparable and nonlinearly separable data. The simulation data results show that SVM has the ability to classify the data better than CT. Ensemble method improve classification performance and more stable than single classifier. This was due to ensemble SVM has the smallest percentage of the average misclassification rate and standard deviation.