Optimization of Intrusion Detection through Fast Hybrid Feature Selection

Existing intrusion detection techniques emphasize on building intrusion detection model based on all features provided. But all features are not relevant and some of them are redundant and useless. In this paper, we propose and investigate a fast hybrid feature selection method - a fusion of Correlation-based Feature Selection, Support Vector Machine and Genetic Algorithm - to determine an optimal feature set. An appropriate feature set helps to build efficient decision model as well as reduced feature set lights up the training and testing process considerably. We have examined the feasibility of our approach by conducting several experiments using KDD 1999 CUP intrusion dataset. Experimental results indicate the reduction of training and testing time by an order of magnitude while maintaining the detection accuracy within tolerable range.