A Novel Feature Extraction Method Assembled with PCA and ICA for Network Intrusion Detection

Abstract

In network intrusion detection systems, feature extraction plays an important role in a sense of improving classification performance and reducing the computational complexity. Principle Component Analysis and Independent Component Analysis are both common feature extraction methods currently. This paper proposed a novel feature extraction method for network intrusion detection and the core of this method is a combiner which is assembled with Principle Component Analysis and Independent Component Analysis. The extracted features are employed by Support Vector Machine (SVM) for classification. The KDDCUP99 data set is used to evaluate the performance of this method. The test results show that the method takes the advantage of PCA and ICA in feature extraction, and has a preferable performance for network intrusion detection.