[1]张 然,潘芷涵,朱 亮,等.基于 PCA 与 WNN 的网络安全态势要素提取方法[J].计算机技术与发展,2023,33(07):119-125.[doi:10. 3969 / j. issn. 1673-629X. 2023. 07. 018]
　ZHANG Ran,PAN Zhi-han,ZHU Liang,et al.An Extraction Method of Network Security Situation Elements Based onPCA and WNN[J].,2023,33(07):119-125.[doi:10. 3969 / j. issn. 1673-629X. 2023. 07. 018]

点击复制

基于 PCA 与 WNN 的网络安全态势要素提取方法()

分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

33

期数:

2023年07期

页码:

119-125

栏目:

网络空间安全

出版日期:

2023-07-10

文章信息/Info

Title:

An Extraction Method of Network Security Situation Elements Based onPCA and WNN

文章编号:

1673-629X(2023)07-0119-07

作者:

张 然; 潘芷涵; 朱 亮; 尹毅峰

郑州轻工业大学 计算机与通信工程学院,河南 郑州 450001

Author(s):

ZHANG Ran; PAN Zhi-han; ZHU Liang; YIN Yi-feng

School of Computer and Communication Engineering,Zhengzhou University of Light Industry,Zhengzhou 450001,China

关键词:

网络安全; 态势要素提取; 主成分分析法; 小波神经网络; 动量因子

Keywords:

network security; situation element extraction; principal component analysis; wavelet neural network; momentum factor

分类号:

TP393

DOI:

10. 3969 / j. issn. 1673-629X. 2023. 07. 018

摘要:

安全态势要素的提取是网络安全态势感知的基础,提取的态势要素质量的好坏直接影响着网络安全态势评估和预测的准确性。 针对大规模网络环境下态势要素提取困难及分类精度不高的问题,提出一种基于 PCA-MF-WNN 的网络安全态势要素提取模型。 该模型利用主成分分析法(PCA) 对预处理后的网络安全数据进行降维,去除冗余的态势要素,然后采用小波神经网络( WNN)对约简后的数据集进行分类训练。 由于传统小波神经网络存在运算效率低和精准度不高的问题,引入动量因子(MF) 对小波函数的伸缩因子、平移因子以及小波神经网络的连接权值进行修正,以提高小波分类器的分类精度与分类效率。 对比实验结果表明,该态势要素提取模型有效提高了态势要素提取的分类精确度和运算效率。

Abstract:

The extraction of situation elements is the basis of network security situation awareness. The quality of situation?

element　extraction directly affects the accuracy of network security situation assessment and prediction. To solve

the problems of difficult situationelement extraction and low classification accuracy in large-scale network environments,　we propose a network security situation elementextraction model based on PCA-MF-WNN. The model uses Principal Component Analysis ( PCA) to reduce the dimension of the prprocessed network security data and remove the redundant situation elements,and then uses Wavelet Neural Network ( WNN) to classifyand train the reduced data sets. Because the traditional wavelet neural network has the problems of low operation efficiency and lowaccuracy,Momentum Factor

?( MF) is introduced to modify the scaling factor,translation factor of a wavelet function,and the connectionweight of the wavelet neural network,so as to improve the classification accuracy and efficiency of wavelet classifier. The results of comparative experiments show that the model effectively improves the classification accuracy and operation efficiency?

of situation element extraction.

常用功能