Universiti Teknologi Malaysia Institutional Repository

A novel method for unsupervised anomaly detection using unlabelled data

Ismail, Abdul Samad and Abdullah, Abdul Hanan and Abu Bak, Kamalrulnizam and Ngadi, Md Asri and Dahlan, Dahliyusmanto and Chimphlee, Witcha (2008) A novel method for unsupervised anomaly detection using unlabelled data. In: Proceedings - The International Conference on Computational Sciences and its Applications, ICCSA 2008 , June 30 - July 3, 2008.

[img] PDF
Restricted to Repository staff only

188Kb

Official URL: http://dx.doi.org/10.1109/ICCSA.2008.70

Abstract

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.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:Network security, intrusion detection, anomaly detection, unsupervised learning, clustering, Principal component analysis.
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:Computer Science and Information System (Formerly known)
ID Code:7467
Deposited By: Maznira Sylvia Azra Mansor
Deposited On:06 Jan 2009 00:25
Last Modified:01 Jun 2010 15:52

Repository Staff Only: item control page