基于聚类与决策树的综合入侵检测算法研究

入侵检测是一种通过实时监测目标系统来发现入侵攻击行为的安全技术,传统的入侵检测系统在有效性、适应性和可扩展性方面都存在着不足。为了使模糊聚类算法获得的聚类结果为全局最优解,改进了传统的模糊C-均值算法,并且在每个聚类的数据集上建立一棵属于该聚类的C4.5决策树,构造了一种新的综合检测算法来确定是否存在入侵。通过实验结果分析,该检测算法降低了误报率,提高了入侵检测的检测性能以及可靠性。

Study of Intrusion Detection Algorithm based on Clustering and Decision Tree

Intrusion detection is a security technology through real-time monitoring system to detect intrusion attacks.Traditional intrusion detection systems have deficiencies in effectiveness,adaptability and scalability.In order to obtain the clustering results for the global optimal solution,paper improves the traditional fuzzy C-means algorithm,creates a data set belonging to the cluster of C4.5 decision tree for each cluster,and Constructs a new integrated detection algorithm to determine whether there is invasion.Through the experimental analysis,detection algorithm reduce the false alarm rate and improve the detection performance and reliability.