Real-Time Computer Network Anomaly Detection Using.pdf

Journal of Advances in Computer Networks, Vol. 1, No. 1, March 2013 Real-Time Computer Network Anomaly Detection Using Machine Learning Techniques Kriangkrai Limthong  as k-nearest neighbor algorithms [7], neural networks [8], Abstract—Detecting a variety of anomalies in computer support vector machines [9], and k-means clustering [10], network, especially zero-day attacks, is one of the real challenges have been applied to detect anomalies. However, most studies for both network operators and researchers. An efficient are batch processing, which collects a certain amount of data technique detecting anomalies in real time would enable before detecting anomalies, and most of these techniques are network operators and administrators to expeditiously prevent therefore not suitable for real-time detection. serious consequences caused by such anomalies. We propose an alternative technique, which based on a combination of time To address this problem, we propose a technique on the series and feature spaces, for using machine learning algorithms basis of time series and feature spaces for real-time anomaly to automatically detect anomalies in real time. Our experimental detection in computer networks. We also present a series of results show that the proposed technique can work well for a experiments that we conducted to examine the performance real network environment, and it is a feasible technique