Semi-supervised anomaly detection in dynamic communication networks

doi:10.1016/j.ins.2021.04.056

	Semi-supervised anomaly detection in dynamic communication networks
	Meng, Xuying 1,2; Wang, Suhang 3; Liang, Zhimin 1; Yao, Di 1; Zhou, Jihua 4; Zhang, Yujun 1
刊名	INFORMATION SCIENCES
	2021-09-01
卷号	571 页码:527-542
关键词	Anomaly detection Semi-supervised learning Generative adversarial networks Self-learning
ISSN号	0020-0255
DOI	10.1016/j.ins.2021.04.056
英文摘要	To ensure the security and stabilization of the communication networks, anomaly detection is the first line of defense. However, their learning process suffers two major issues: (1) inadequate labels: there are many different kinds of attacks but rare abnormal nodes in mt of these atstacks; and (2) inaccurate labels: considering the heavy network flows and new emerging attacks, providing accurate labels for all nodes is very expensive. The inadequate and inaccurate label problem challenges many existing methods because the majority normal nodes result in a biased classifier while the noisy labels will further degrade the performance of the classifier. To tackle these issues, we propose SemiADC, a Semi-supervised Anomaly Detection framework for dynamic Communication networks. SemiADC first approximately learns the feature distribution of normal nodes with regularization from abnormal ones. It then cleans the datasets and extracts the nodes sasainaccurate labels by the learned feature distribution and structure-based temporal correlations. These self-learning processes run iteratively with mutual promotion, and finally help increase the accuracy of anomaly detection. Experimental evaluations on real-world data sets demonstrate the effectiveness of our SemiADC, which performs substantially better than the state-of-art anomaly detection approaches without the demand of adequate and accurate supervision. (c) 2021 Published by Elsevier Inc.
资助项目	National Key Research and Development Program of China[2018YFB1800403] ; National Science Foundation of China[61902382] ; National Science Foundation of China[61972381] ; National Science Foundation of China[61672500] ; Strategic Priority Research Program of Chinese Academy of Sciences[XDC02030500] ; research program of Network Computing Innovation Research Institute[E061010003]
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER SCIENCE INC
WOS记录号	WOS:000683548900009
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/17231]
专题	中国科学院计算技术研究所
通讯作者	Zhang, Yujun
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 2.Purple Mt Labs, Nanjing, Peoples R China 3.Penn State Univ, Coll Informat Sci & Technol, State Coll, PA USA 4.Jin Mei Commun, Chogqing, Peoples R China
推荐引用方式 GB/T 7714	Meng, Xuying,Wang, Suhang,Liang, Zhimin,et al. Semi-supervised anomaly detection in dynamic communication networks[J]. INFORMATION SCIENCES,2021,571:527-542.
APA	Meng, Xuying,Wang, Suhang,Liang, Zhimin,Yao, Di,Zhou, Jihua,&Zhang, Yujun.(2021).Semi-supervised anomaly detection in dynamic communication networks.INFORMATION SCIENCES,571,527-542.
MLA	Meng, Xuying,et al."Semi-supervised anomaly detection in dynamic communication networks".INFORMATION SCIENCES 571(2021):527-542.