An Effective Distance-Computing Method for Network Anomaly Detection

摘要: Currentlymany traditional network anomaly detection algorithms are proposed to distinguish anomalies from heavy traffic. However, most of them based on data mining or machine learning methods, which brings unexpected computational cost and high false alarm rates. In this paper, we propose a simple distance-computing algorithm for detection, is able normal traffic using but effective mechanism. Experimental results the well-known KDD Cup 1999 dataset demonstrate it can effectively detect with true positives, low positives acceptable cost.