A Novel Method for Unsupervised Anomaly Detection Using  Unlabelled Data

Most current intrusion detection methods cannot process large amounts of audit data for real-time operation. In this paper, anomaly network intrusion detection method based on principal component analysis (PCA) for data reduction and fuzzy adaptive resonance theory (fuzzy ART) for classifier is presented. Moreover, PCA is applied to reduce the high dimensional data vectors and distance between a vector and its projection onto the subspace reduced is used for anomaly detection. Using a set of benchmark data from KDD (knowledge discovery and data mining) competition designed by DARPA for demonstrate to detection intrusions. Experimental results show the proposed model can classify the network connections with satisfying performance.