###
DOI:
计算机系统应用英文版:2012,21(8):69-72,31
本文二维码信息
码上扫一扫!
基于AdaBoost 的入侵检测技术探索与分析
(渭南师范学院 数学与信息科学学院, 渭南 714000)
Intrusion Detection Technology Based on AdaBoost
(Weinan Normal University, Mathematics and the Information Science Institute, Weinan 714000, China)
摘要
图/表
参考文献
相似文献
本文已被:浏览 1702次   下载 2780
Received:November 23, 2011    Revised:March 05, 2012
中文摘要: 阐明了入侵检测系统的监测过程,提出在入侵检测的分析方法中通过AdaBoost 框架的循环迭代,在每次迭代中,由该算法产生一个带权值的分类器,迭代结束产生多个分类器,最后将这些分类器进行加权联合,得到一个具有较高识别率的分类器,进而克服采用单一分类算法产生的识别率难以满足系统要求的缺陷,从而达到系统对攻击识别率提高,误警率降低的目的,以KDD99 作为实验样本数据源,仿真实验表明该方法检测预警准确率高。
中文关键词: 样本特征  分类器  入侵检测  权值  分类决策
Abstract:This paper illuminates the intrusion detection system monitoring process and puts forward that in intrusion detection methods of analysis by the iterative AdaBoost framework, in each iteration, the algorithm produces a belt of the weight values classifier, the iterative end into multiple classifier. Finally, the classifier are weighted joint to get a higher rate of classifier, and hence overcome classification algorithm USES a single produced to meet the requirements of the recognition system defect, so as to improve the system to attack rate,reduce false alarm rate of purpose, in KDD99 were selected as the experimental data. The simulation experiments show that the method is accurate in early warning detection.
文章编号:     中图分类号:    文献标志码:
基金项目:陕西省自然科学基金(2011JM8020);渭南市自然科学基金(2011KYJ-1)
引用文本:
阴国富.基于AdaBoost 的入侵检测技术探索与分析.计算机系统应用,2012,21(8):69-72,31
YIN Guo-Fu.Intrusion Detection Technology Based on AdaBoost.COMPUTER SYSTEMS APPLICATIONS,2012,21(8):69-72,31