An Efficient Approach for Intrusion Detection in Reduced Features of KDD99 Using ID3 and Classification with KNNGA

KDDCUP 1999 Dataset widely used dataset of data mining in the field of intrusion detection by various researchers. This dataset are publicly available for the users. Intrusion detection is the key challenges for the users because the intrusion may corrupt or destroy the network services. The intrusion detection system is classified into two categories: Network based intrusion detection system and Misuse intrusion detection system. In this paper, novel method is for intrusion detection with feature reduction using partially ID3 algorithm to find higher information gain for attribute selection and KNN based GA (genetic algorithm) is applied for classification and detection of intrusions on KDD dataset. The simulation & analysis of the method is done on MATLAB2012A. The experimental scenario of proposed methodology produces better result when it compared with some existing approaches, for the measurement of the result comparing with the different performance metrics parameters such as sensitivity, specificity and accuracy.