Significant of features selection for detecting network intrusions

Intrusion Detection System (IDS) is used to identify unknown or new type of attacks especially in dynamic environments as business and mobile networks. For that importance, IDS has become one of targeted research area that focuses on information security. Among different techniques, Steady State Genetic-Based Machine Learning Algorithm (SSGBML) offers the ability to detect intrusions especially in changing environments. The objective of this paper is to incorporate several feature selection. Selection network features has a great importance to increase detection rate, which is itself a problem in Intrusion Detection System (IDS). Since elimination of the insignificant and/or useless features leads to a simplified problem and enhance detection rate. By combining different selected features that will be evaluated, where this will lead us to determine suitable combination features to attain best results. In SSGBML, Zeroth Level Classifier System (ZCS) plays the role of detector by matching incoming environment message with classifiers to determine whether it is normal or intrusion. The experiments and evaluations for compound methods were performed on KDD 99 dataset to detect network intrusions.