Intrusion detection based on Core Vector Machine and ensemble classification methods

With the widespread use of Internet, the possibilities of exposing confidential data to invaders or attackers increases. Intrusion Detection System (IDS) is used for detecting various intrusions in network environment and to prevent data from malicious attackers. In this paper, a combined algorithm based on Principal Component Analysis (PCA) and Core Vector Machine (CVM), which is an extremely fast classifier, is proposed for intrusion detection. PCA is used as feature extraction technique to select principal features from the intrusion detection KDDCup´99 dataset and an intrusion detection model is constructed by CVM algorithm. The effectiveness of the features selected is also tested on ensemble based classifiers and the results are compared with the standard classifiers.