PRACTICAL NETWORK ANOMALY DETECTION USING DATA MINING TECHNIQUES
Xiejun Ni, Daojing He, Farooq Ahmad · VFAST Transactions on Software Engineering · 2016
Network anomaly detection is an effective way to detect intrusions which defends our computer systems or network from attackers on the Internet. In this paper, we introduce the current research works in network anomaly detection and consider serveral pratical solutions for this issue. Different from signature-based method, data mining techniques can automatically extract normal pattern from a large set of network data and distinguish them from each other. However, those data mining techniques, such as classification, clustering, association rules and feature selection, can not be applied into this problem directly due to the characteristic of network data and technique themseleves. We analyze those unfitness and propose some adaptation to detect anomaly timely and accurately.