An ensemble feature selection approach using hybrid kernel based SVM for network intrusion detection system

Feature selection is a process of identifying relevant feature subset that leads to the machine learning algorithm in well-defined manner. In this paper, anovel ensemble approach comprises Relief Attribute Evaluation and hybrid kernel-based support vector (HK-SVM) proposed as method for network intrusion detection system (NIDS). A Hybrid along with combination Gaussian Polynomial methods used kernel (SVM). The key issue select yields good accuracy at minimal computational cost. implemented compared classical SVM simple kernel. Kyoto2006+, bench mark dataset,is experimental evaluation then observations are drawn.