数据清洗下的改进半监督聚类入侵检测算法研究  

A Data Cleaning Based on Improved Semi-supervised Clustering Algorithm for Intrusion Detection System

作　　者：周志平[1] 陈晓洁[1] ZHOU Zhiping CHEN Xiaojie(School of Information engineering, LongYan University, Longyan, Fujian 36400)

机构地区：[1]龙岩学院信息工程学院,福建龙岩364000

出　　处：《武夷学院学报》2017年第3期67-71,共5页Journal of Wuyi University

基　　金：龙岩学院青年教师攀登项目(LQ2013008;LQ2014001)

摘　　要：针对半监督聚类算法易受噪点的影响,提出一种基于数据清洗的改进半监督聚类算法DCSC。将噪点从数据集中提取出来另行分析。从数据清洗处理后的数据集中抽取若干正常与异常样本分别计算作为初始样本辅助K-Means算法进行聚类。实验结果表明,与现有相关算法相比,该算法具有检测未知攻击的能力,且具有更高的攻击检测率以及更低的误报率。A data cleaning based improved semi-supervised clustering algorithm is proposed in this paper in order to overcome the problem for semi-supervised clustering algorithm noise sensitivity. Noise sample in the data sets will be extracted for further analysis.From the data cleaned data sets, we collect samples which are labled normal and abnormal, then calculate them respectively to support the initialization phase of K-Means clustering algorithm. Experimental results show that proposed clustering algorithm has the ability to detect unknown attacks, and has a higher attack detection rate and lower false positive rate.

关 键 词：数据清洗 半监督聚类 入侵检测 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]