An Intelligent Network Intrusion Detection and Prevention System

With the wide use of Internet and network connectivity, it is important to prevent unauthorized access to system resources and data. In this work, an intelligent network intrusion detection and prevention system is presented for detecting and preventing network attacks that incorporates a knowledge based system and data mining techniques. Hybrid data mining process model is followed for data mining tasks to extract hidden knowledge from KDDCup¿99 intrusion dataset. J48 decision tree, JRip rule induction, Naïve Bayes and Multilayer Perceptron (MLP) Neural Network are adopted to construct a predictive model on total datasets of 63, 661 instances. This work supports network administrators to fill the knowledge gap they have to detect and prevent network attacks efficiently and effectively. Experimental result shows that, the proposed system performs 91.43 percent and 83 percent detection accuracy and user acceptance, respectively. The system cannot update the knowledge of prevention techniques automatically which need further researches.