A COMPARATIVE STUDY OF RULE BASED CLASSIFIER AND DECISION TREE IN MACHINE LEARNING

Abstract

In Machine Learning, Decision Tree is an effective method of Classification. In this paper, the two methods Rule based classifier and Decision tree have been used to generate rules and compare the accuracies and precisions. As a result, it is found that decision tree classifier gives the good accuracy. The metrics entropy and information gain are used for finding the best splitting attribute for inducing tree of Iris dataset. Also, confusion matrix is used for finding accuracy of both the methods. Our experimental results show that Decision tree has excellent performance in terms of accuracy and error rates. Also, we find that confidence interval for accuracy of Decision Tree becomes tighter than Rule based classifier on increasing the number of data.