An intelligent intrusion detection system using genetic based feature selection and Modified J48 decision tree classifier

In this paper, intelligent algorithms for intrusion detection are proposed which detect the network attacks as normal or anomaly based attacks by performing effective preprocessing and classification. This system uses a new genetic algorithm approach for pre-processing and Modified J48 classification algorithm to identify the intended activities. The new genetic based feature selection algorithm proposed in this paper is helpful to identify the important features needed to classify the normal and anomaly records. The proposed intelligent IDS has been empirically tested in a simulated environment and the experimental results show that the proposed method provides higher detection accuracy than the existing methods in terms of detection rate with reduced false rate. The salient contributions of this paper are, the proposed a new processing technique for removing noisy data in the KDD cup´99 dataset, identification of the optimal features selection by applying modified genetic algorithm and finally the proposed of modified J48 decision tree algorithm for efficient classification for providing intelligent network intrusion detection.