Classifier Selection Model Based on Gain Ratio Feature Selection Method

The computer networks usage has grown enormous and widespread, which has increased the number of new threats to a great extent. Intruder is one of the most publicized threats to security. In recent years, intrusion detection has emerged as an important technique for network security. Although there are some existing techniques for intrusion detection, there is a need to improve the performance. Data mining techniques have been applied as a new approach for intrusion detection. The quality of the feature selection methods is one of the important factors that affect the effectiveness of the Intrusion Detection system (IDS). In this paper, feature selection method, Gain Ratio is used to extract an optimal subset of features, which are then subjected to a set of classification algorithms to analyze KDDCup'99 dataset. We used 10-fold cross validation for building our proposed model. The classification algorithms are compared in terms of accuracy, detection rate, false alarm rate and time taken.