基于改进K-Means算法的入侵检测方法
首发时间:2017-02-17
摘要:近年来数据挖掘技术在入侵检测领域的应用越来越多,K-Means算法是聚类算法中一种高效的划分算法,应用广泛,但是基于K-Means聚类算法的入侵检测系统仍然存在检测率低和误报率高的不足。针对此问题,本文提出一种将改进K-Means算法和FP-Growth算法相结合(KMFP)的入侵检测方法,将K-Means改进算法的分类特性和FP-Growth算法模式挖掘相结合,降低系统的误报率,同时保证系统的检测率和检测的效率。利用KDDCup99数据集进行实验,结果表明KMFP方法能有效降低误报率,同时保证系统的检测率和检测的效率。
关键词: 入侵检测 聚类分析 K-Means算法 FP-Growth算法
For information in English, please click here
Intrusion Detection Method Based On Improved K-Means Algorithm
Abstract:Recent years, more and more data mining technology are applied in the field of intrusion detection s, K-Means is an efficient division algorithm in clustering. However, the existing intrusion detection system based on K-Means algorithm still has the shortage of having low detection rate and high false alarm rate. To solve this problem, this paper proposed a method for intrusion detection based on improved K-Means algorithm and FP-Growth algorithm. The Classification features of K-Means can be well combined with FP-Growth which reduce the system's false alarm rate, and at the same time ensure the detection rate of the system and the efficiency of the system. KDDCup99 dataset was used for the experiment. The experimental results show that the KMFP method can effectively reduce the false alarm rate, also the KMFP method can guarantee the detection rate and the efficiency of detection at the same time.
Keywords: intrusion detection clustering K-Means algorithm FP-Growth algorithm
基金:
论文图表:
引用
No.4717129117256914****
同行评议
共计0人参与
勘误表
基于改进K-Means算法的入侵检测方法
评论
全部评论0/1000