基于有监督判别投影的网络安全数据降维算法

doi:10.11959/j.issn.1000-436x.2021117

通信学报 ›› 2021, Vol. 42 ›› Issue (6): 84-93.doi: 10.11959/j.issn.1000-436x.2021117

基于有监督判别投影的网络安全数据降维算法

郭方方, 吕宏武, 任威霖, 王瑞妮

哈尔滨工程大学计算机科学与技术学院，黑龙江哈尔滨 150001

修回日期:2021-03-01 出版日期:2021-06-25 发布日期:2021-06-01
作者简介:郭方方（1973− ），男，黑龙江哈尔滨人，博士，哈尔滨工程大学副教授、硕士生导师，主要研究方向为计算机网络应用、新型网络体系结构、网络安全态势感知、云监控等
吕宏武（1983− ），男，山东日照人，博士，哈尔滨工程大学副教授、博士生导师，主要研究方向为网络安全、移动云计算与移动边缘计算、形式化建模与性能评价等
任威霖（1997− ），男，黑龙江哈尔滨人，哈尔滨工程大学硕士生，主要研究方向为网络数据预处理、网络态势感知预测等
王瑞妮（1994− ），女，山西运城人，哈尔滨工程大学硕士生，主要研究方向为流形学习、网络数据异常处理等
基金资助:
国家自然科学基金资助项目(61872104);中央高校基础科研业务费专项资金资助项目(3072020CF0603)

Reduction algorithm based on supervised discriminant projection for network security data

Fangfang GUO, Hongwu LYU, Weilin REN, Ruini WANG

College of Computer Science and Technology, Harbin Engineering University, Harbin 150001, China

Revised:2021-03-01 Online:2021-06-25 Published:2021-06-01
Supported by:
The National Natural Science Foundation of China(61872104);The Fundamental Research Fund for the Central Universities(3072020CF0603)

摘要/Abstract

摘要：

针对传统流形学习在数据降维时不考虑原数据类别和聚类程度低的缺陷，提出了一种有监督判别投影（SDP）的流形学习降维算法来改善网络安全数据降维效果。在近邻矩阵基础上，利用数据集的类别标签信息，构建有监督判别矩阵，变无监督流形学习为有监督学习，寻找一个同时具有最大全局散度矩阵和最小局部散度矩阵的低维投影子空间，保证了降维投影后同类数据聚集而异类数据分散的特性。实验结果显示，与传统降维算法相比，所提算法可以较低的时间复杂度去除冗余数据，并且降维后的数据聚类效果更好，异类样本更分散，适用于实际的网络安全数据分析模型。

关键词: 数据降维, 流形学习, 有监督学习, 判别投影

Abstract:

In response to the problem that for dimensionality reduction, traditional manifold learning algorithm did not consider the raw data category information, and the degree of clustering was generally at a low level, a manifold learning dimensionality reduction algorithm with supervised discriminant projection (SDP) was proposed to improve the dimensionality reduction effects of network security data.On the basis of the nearest neighbor matrix, the label information of the raw data category was exploited to construct a supervised discriminant matrix in order to translate unsupervised popular learning into supervised learning.The target was to find a low dimensional projective space with both maximum global divergence matrix and minimum local divergence matrix, ensuring that the same kind of data was concentrated and heterogeneous data was scattered after dimensionality reduction projection.The experimental results show that the SDP algorithm, compared with the traditional dimensionality reduction algorithms, can effectively remove redundant data with low time complexity.Meanwhile the data after dimensionality reduction is more concentrated, and the heterogeneous samples are more dispersed, suitable for the actual network security data analysis model.

Key words: data dimension reduction, manifold learning, supervised learning, discriminant projection

中图分类号:

TP309.2

郭方方, 吕宏武, 任威霖, 王瑞妮. 基于有监督判别投影的网络安全数据降维算法[J]. 通信学报, 2021, 42(6): 84-93.

Fangfang GUO, Hongwu LYU, Weilin REN, Ruini WANG. Reduction algorithm based on supervised discriminant projection for network security data[J]. Journal on Communications, 2021, 42(6): 84-93.

图/表 9

图1

表1

图2

图3

图4

表2

图5

表3

图6

参考文献 23

[1]	WANG Z , PARKINSON T , LI P X ,et al. The squeaky wheel:machine learning for anomaly detection in subjective thermal comfort votes[J]. Building and Environment, 2019,151: 219-227.
[2]	VIKRAM M , PAVAN R , DINESHBHAI N D ,et al. Performance evaluation of dimensionality reduction techniques on high dimensional data[C]// 2019 3rd International Conference on Trends in Electronics and Informatics. Piscataway:IEEE Press, 2019: 1169-1174.
[3]	BYRNE J J , MORGAN J L , TWICKLER D M ,et al. Utility of follow-up standard sonography for fetal anomaly detection[J]. American Journal of Obstetrics and Gynecology, 2020,222(6): 615.e1-615.e9.
[4]	NAKAZAWA T , KULKARNI D V . Anomaly detection and segmentation for wafer defect patterns using deep convolutional encoder–decoder neural network architectures in semiconductor manufacturing[J]. IEEE Transactions on Semiconductor Manufacturing, 2019,32(2): 250-256.
[5]	TERAUCHI T , AIKEN A . Secure information flow as a safety problem[M]. Berlin: Springer, 2005: 352-367.
[6]	DURUMERIC Z , MA Z N , SPRINGALL D ,et al. The security impact of HTTPS interception[C]// 2017 Network and Distributed System Security Symposium. Virginia:the Internet Society, 2017: 1-5.
[7]	WANG L L , . Research on distributed parallel dimensionality reduction algorithm based on PCA algorithm[C]// 2019 IEEE 3rd Information Technology,Networking,Electronic and Automation Control Conference. Piscataway:IEEE Press, 2019: 1363-1367.
[8]	GHOSH J , SHUVO S B . Improving classification model's performance using linear discriminant analysis on linear data[C]// 2019 10th International Conference on Computing,Communication and Networking Technologies. Piscataway:IEEE Press, 2019: 1-5.
[9]	WU D , XIONG N X , HE J R ,et al. Critical data points-based unsupervised linear dimension reduction technology for science data[J]. The Journal of Supercomputing, 2016,72(8): 2962-2976.
[10]	SARASWATI A , NGUYEN V T , HAGENBUCHNER M ,et al. High-resolution self-organizing maps for advanced visualization and dimension reduction[J]. Neural Networks, 2018,105: 166-184.
[11]	张军平, 王珏 . 主曲线研究综述[J]. 计算机学报, 2003,26(2): 129-146.
	ZHANG J P , WANG J . An overview of principal curves[J]. Chinese Journal of Computers, 2003,26(2): 129-146.
[12]	BALCAN M F , LIANG Y , SONG L . Communication efficient distributed kernel principal component analysis[J]. Computer Science, 2016,27(4): 555-559.
[13]	NOURI M , MIVEHCHY M , AGHDAM S A . Adaptive time-frequency kernel local fisher discriminant analysis to distinguish range deception jamming[C]// 2015 6th International Conference on Computing,Communication and Networking Technologies. Piscataway:IEEE Press, 2015: 1-5.
[14]	CAO Z Y , JI G L , TAN C . Improvement of algorithm multi-manifold LLE learning[J]. Computer Engineering and Applications, 2018,54(24): 156-163.
[15]	石陆魁, 郭林林, 房子哲 ,等. 基于Spark的并行ISOMAP算法[J]. 中国科学技术大学学报, 2019,49(10): 842-850.
	SHI L K , GUO L L , FANG Z Z ,et al. Parallel ISOMAP algorithm based on Spark[J]. Journal of University of Science and Technology of China, 2019,49(10): 842-850.
[16]	SUN W W , YANG G , DU B ,et al. A sparse and low-rank near-isometric linear embedding method for feature extraction in hyperspectral imagery classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017,55(7): 4032-4046.
[17]	CHEN X Z . LTSA algorithm for dimension reduction of microarray data[J]. Advanced Materials Research, 2013,645: 192-195.
[18]	ZHU B , LIU J Z , CAULEY S F ,et al. Image reconstruction by domain-transform manifold learning[J]. Nature, 2018,555(7697): 487-492.
[19]	DING C , QI H D . Convex optimization learning of faithful Euclidean distance representations in nonlinear dimensionality reduction[J]. Mathematical Programming, 2017,164(1/2): 341-381.
[20]	NING X , LI W J , TANG B ,et al. BULDP:biomimetic uncorrelated locality discriminant projection for feature extraction in face recognition[J]. IEEE Transactions on Image Processing, 2018,27(5): 2575-2586.
[21]	VURAL E , GUILLEMOT C . A study of the classification of low-dimensional data with supervised manifold learning[J]. The Journal of Machine Learning Research, 2017,18(1): 5741-5795.
[22]	李锋, 汤宝平, 王家序 ,等. 基于图嵌入概率半监督判别分析的故障辨识[J]. 机械工程学报, 2017,53(9): 92-100.
	LI F , TANG B P , WANG J X ,et al. Fault identification method based on graph-implanted probability-based semi-supervised discriminant analysis[J]. Journal of Mechanical Engineering, 2017,53(9): 92-100.
[23]	GURUNG S , GHOSE M K , SUBEDI A . Deep learning approach on network intrusion detection system using NSL-KDD dataset[J]. International Journal of Computer Network and Information Security, 2019,11(3): 8-14.

标记类型	训练集数目	测试集数目
Normal	67 343	13 449
DoS	45 927	9 234
PROBE	11 656	2 289
R2L	995	209
U2R	52	11

数据规模	运行时间/s
	线性算法	流形学习算法
	PCA算法	LE算法	LDP算法	SDP算法
300	0.031	0.08	0.37	0.38
600	0.072	0.17	1.21	1.31
1 200	0.075	0.61	2.93	3.27
2 400	0.075	1.63	9.21	9.72
4 800	0.075	4.19	32.51	33.21
9 600	0.138	12.41	110.31	90.53
19 200	0.138	30.36	200.24	130.48

数据规模		综合性能指数
数据规模	LE	LDP	SDP
300	61.13	75.57	111.16
600	57.53	46.21	64.49
1 200	42.52	38.17	51.67
2 400	24	24.29	34.77
4 800	18.67	13.76	20.35
9 600	12.61	8.11	14.93
19 200	10.31	8.94	20.72

基于有监督判别投影的网络安全数据降维算法

Reduction algorithm based on supervised discriminant projection for network security data

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 23

相关文章 1

Metrics

推荐阅读 0