An enhanced support vector machine model for intrusion detection

摘要: Design and implementation of intrusion detection systems remain an important research issue in order to maintain proper network security. Support Vector Machines (SVM) as a classical pattern recognition tool have been widely used for detection. However, conventional SVM methods do not concern different characteristics features building system. We propose enhanced model with weighted kernel function based on the training data Rough set theory is adopted perform feature ranking selection task new model. evaluate KDD dataset UNM dataset. It suggested that proposed outperformed precision, computation time, false negative rate