A Robust Feature Normalization Scheme and an Optimized Clustering Method for Anomaly-Based Intrusion Detection System

摘要: Intrusion detection system(IDS) has played a central role as an appliance to effectively defend our crucial computer systems or networks against attackers on the Internet. Traditional IDSs employ signature-based methods anomaly-based which rely labeled training data. However, they have several problems, for example, it consumes huge amounts of cost and time acquire data, often experienced difficulty in detecting new types attack. In order cope with many researchers proposed various kinds algorithms years. Although do not require data capability detect unforeseen attacks, are based assumption that ratio attack normal is extremely small. may be satisfied realistic situation because some most notably denial-of-service consist large number simultaneous connections. Consequently if fails, performance algorithm will deteriorate. this paper, we present normalization clustering method can overcome limitation We evaluated using KDD Cup 1999 set. Evaluation results show approach constant irrespective increase ratio.