Rank Allocation to J48 Group of Decision Tree Classifiers using Binary and Multiclass Intrusion Detection Datasets

Abstract

The wide acceptance of decision tree classifiers lies with their fast performance and simple nature. The J48 group of decision tree classifiers are widely used for classification and decision-making process. In this paper, three popular J48 group classifiers, namely J48, J48Consolidated and J48Graft are evaluated using both binary and multi-class datasets across thirteen performance matrices, which is unique in its area. In order to come across a versatile classifier, the evaluated results of these classifiers are nourished to a prominent multi-criteria decision-making module called Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) for suitable rank allocation.