The Improvement of Cluster Analysis in Intrusion Detection System

In view of the shortcomings of the traditional clustering algorithm in intrusion detection system, this paper proposes a method of selecting the initial clustering center based on density, which can overcome the problem of K value in ordinary K-Means. The improved intrusion detection model can achieve good clustering effect. Compared with the traditional K-Means, it is found that the improved algorithm can obtain higher detection rate and lower false alarm rate than the traditional K-Means.