Grading Method of Ensemble and Genetic Algorithm for Intrusion Detection System

Abstract

Intrusion Detection System is very important tool for network security. However, Intrusion Detection System suffers from the problem of handling large volume of data and produces high false positive rate. In this paper, a novel Grading method of ensemble has proposed to overcome limitation of intrusion detection system. Partial decision tree (PART), RIpple DOwn Rule (RIDOR) learner and J48 decision tree have used as base classifiers of Grading classifier. Optimzed Genetic Search algorithm have used for selection of features.These three base classifiers have graded using RandomForest decision tree as a Meta classifier. Experimental results show that the proposed Grading method of classification offers accuracies of 81.3742%, 99.9159% and 99.8023% on testing, training datasets and cross validation respectively. It is found that the proposed graded classifier outperform its base classifiers and existing hybrid intrusion detection system in term of accuracy, false positive rate and model building time.