An approach towards intrusion detection using PCA feature subsets and SVM

Presently many intrusion detection approaches are available but have drawbacks like training overhead as well as their performance factor. Increased detection rate with less false alarms can enhanced the efficiency of an intrusion detection system. One of the main limitations is the processing of raw features for classification which increases the architecture complexity and decreases the accuracy of detecting intrusions. Because of the limitations in earlier approaches, this PCA based subsets has been proposed. An SVM based IDS mechanism with Principal Component Analysis (PCA) feature subsets has been presented. Support Vector Machines (SVM) used as classifier to test and train the subsets of extracted features with Radial Basis Function (RBF) kernel.