Deep graph level anomaly detection with contrastive learning

doi:10.1038/s41598-022-22086-3

CSpace > 应用数学研究所

	Deep graph level anomaly detection with contrastive learning
	Luo, Xuexiong 1,5; Wu, Jia 1; Yang, Jian 1; Xue, Shan 2; Peng, Hao 3; Zhou, Chuan4 ; Chen, Hongyang 5; Li, Zhao 5,6; Sheng, Quan Z.1
	2022-11-18
发表期刊	SCIENTIFIC REPORTS
ISSN	2045-2322
卷号	12 期号:1 页码:11
摘要	Graph level anomaly detection (GLAD) aims to spot anomalous graphs that structure pattern and feature information are different from most normal graphs in a graph set, which is rarely studied by other researchers but has significant application value. For instance, GLAD can be used to distinguish some different characteristic molecules in drug discovery and chemical analysis. However, GLAD mainly faces the following three challenges: (1) learning more comprehensive graph level representations to differ normal graphs and abnormal graphs, (2) designing an effective graph anomaly evaluation paradigm to capture graph anomalies from the local and global graph perspectives, (3) overcoming the number imbalance problem of normal and abnormal graphs. In this paper, we combine graph neural networks and contrastive learning to build an end-to-end GLAD framework for solving the three challenges above. We aim to design a new graph level anomaly evaluation way, which first utilizes the contrastive learning strategy to enhance different level representations of normal graphs from node and graph levels by a graph convolution autoencoder with perturbed graph encoder. Then, we evaluate the error of them with corresponding representations of the generated reconstruction graph to detect anomalous graphs. Extensive experiments on ten real-world datasets from three areas, such as molecular, protein and social network anomaly graphs, show that our model can effectively detect graph level anomaly from the majority and outperform existing advanced methods.
DOI	10.1038/s41598-022-22086-3
收录类别	SCI
语种	英语
资助项目	ARC DECRA Project[DE200100964] ; NSFC[61872360] ; NSFC[62271452] ; CAS Project for Young Scientists in Basic Research[YSBR-008] ; Key Research Project of Zhejiang Lab[2022PI0AC01] ; National Key R&D Program of China[2022YFB4500300]
WOS研究方向	Science & Technology - Other Topics
WOS类目	Multidisciplinary Sciences
WOS记录号	WOS:000885379900026
出版者	NATURE PORTFOLIO
引用统计
文献类型	期刊论文
条目标识符	http://ir.amss.ac.cn/handle/2S8OKBNM/60656
专题	应用数学研究所
通讯作者	Wu, Jia
作者单位	1.Macquarie Univ, Sch Comp, Sydney, NSW, Australia 2.Univ Wollongong, Sch Comp & Informat Technol, Wollongong, NSW, Australia 3.Beihang Univ, Beijing Adv Innovat Ctr Big Data & Brain Comp, Beijing, Peoples R China 4.Acad Math & Syst Sci, Chinese Acad Sci, Beijing, Peoples R China 5.Zhejiang Lab, Hangzhou, Peoples R China 6.Hangzhou Yugu Technol, Hangzhou, Peoples R China
推荐引用方式 GB/T 7714	Luo, Xuexiong,Wu, Jia,Yang, Jian,et al. Deep graph level anomaly detection with contrastive learning[J]. SCIENTIFIC REPORTS,2022,12(1):11.
APA	Luo, Xuexiong.,Wu, Jia.,Yang, Jian.,Xue, Shan.,Peng, Hao.,...&Sheng, Quan Z..(2022).Deep graph level anomaly detection with contrastive learning.SCIENTIFIC REPORTS,12(1),11.
MLA	Luo, Xuexiong,et al."Deep graph level anomaly detection with contrastive learning".SCIENTIFIC REPORTS 12.1(2022):11.