考虑到入侵检测问题中样本数据的不平衡性,可以将单分类模型应用到入侵检测中.通过应用贝叶斯方法,对单分类支持向量机模型进行改进,使之成为一种概率模型,从而更加符合入侵过程的随机特性;应用主成分分析技术对原始数据在各个方向上进行等方差处理,使之更加符合该模型对数据正态分布的前提假设.同时,在模型求解上采用分治的思想,对数据进行分组计算,实现对大数据的高效求解.在标准入侵检测数据集NSL-KDD上进行测试,实验结果达到87.46％的检测准确率,从而验证了该方法可以在入侵检测中得到有效利用.
[1] 黄谦, 王震, 韦韬, 陈昱, HUANG Qian, WANG Zhen, WEI Tao, CHEN Yu. 基于One-class SVM的实时入侵检测系统 [J]. 计算机工程 2006.doi:10.3969/j.issn.1000-3428.2006.16.048
[2] 包潘晴, 杨明福, Bao Panqing, Yang Mingfu. 基于KPCA和SVM的网络入侵检测 [J]. 计算机应用与软件 2006.doi:10.3969/j.issn.1000-386X.2006.02.050
[3] 田俊峰, 刘涛, 陈小祥, TIAN Jun-feng, LIU Tao, CHEN Xiao-xiang. 入侵检测系统的评估方法与研究 [J]. 计算机工程与应用 2008.doi:10.3778/j.issn.1002-8331.2008.09.034
[4] ALIREZA Ghasemi, HAMID R Rabiee, MOHAMMAD T Manzuri. A Bayesian approach to the data description problem [A]. California:AAAI Press 2012.
[5] DOROTHY E D. An intrusion detection model [J]. {H}IEEE Transactions on Software Engineering 1987, 13(2).
[6] SCHLKOPF B, PLATT J C, TAYLOR J S. Estimating the support of a high dimensional distribution [J]. {H}Neural Computation 200c, 13(7).
[7] DAVID M.J. TAX, ROBERT P.W. DUIN. Support Vector Data Description [J]. Machine learning 2004, 1(1).
[8] SHEHROZ S Khan, MICHAEL G Madden. A survey of recent trends in one class classication [J]. Artificial Intelligence and Cognitive Science 2010, 45(16).
[9] PORTNOY L, ESKIN E, STOLFO S. Intrusion detection with unlabeled data using clustering [A]. Philadelphia:[s.n.] 2001.
[10] BEN H A, GUYON I. Detecting stable clusters using principal component analysis [A]. 2003.
[11] 许国根, 贾瑛. 模式识别与智能计算的MATLAB实现 [M]. {H}北京:北京航空航天大学出版社 2012.
[12] ISABELLE GUYON, ANDRE ELISSEEFF. An introduction to variable and feature selection [J]. {H}JOURNAL OF MACHINE LEARNING RESEARCH 2003, 3.
[13] YE Nong, LI Xiangyang, CHEN Qiang. Probabilistic techniques for intrusion detection based on computer audit data [J]. {H}IEEE Transactions on Systems Man and Cybernetics 2001, 31(4).
[14] 边肇琪, 张学工. 模式识别 [M]. {H}北京:清华大学出版社 2000.
[15] CHANG Chihchung, LIN Chihjen. LIBSVM:A Library for Support Vector Machines [DB/OL]. http://www.csie.ntu.edu.tw/～cjlin/papers/libsvm.pdf 2013.
[16] MAHBOD Tavallaee, EBRAHIM Bagheri, WEI Lu. A detailed analysis of the KDD CUP 99 Data Set [A]. [s.n.] 2009.
[17] VICTORIA J. HODGE, JIM AUSTIN. A Survey of Outlier Detection Methodologies [J]. Artificial Intelligence Review: An International Science and Engineering Journal 2004, 2(2).
