CGCNImp: a causal graph convolutional network for multivariate time series imputation

doi:10.7717/peerj-cs.966

	CGCNImp: a causal graph convolutional network for multivariate time series imputation
	Liu, Caizheng 1,2; Cui, Guangfan 2; Liu, Shenghua 1
	2022-04-29
发表期刊	PEERJ COMPUTER SCIENCE
卷号	8 页码:23
摘要	Background. Multivariate time series data generally contains missing values, which can be an obstacle to subsequent analysis and may compromise downstream applications. One challenge in this endeavor is the presence of the missing values brought about by sensor failure and transmission packet loss. Imputation is the usual remedy in such circumstances. However, in some multivariate time series data, the complex correlation and temporal dependencies, coupled with the non-stationarity of the data, make imputation difficult. Methods. To address this problem, we propose a novel model for multivariate time series imputation called CGCNImp that considers both correlation and temporal dependency modeling. The correlation dependency module leverages neural Granger causality and a GCN to capture the correlation dependencies among different attributes of the time series data, while the temporal dependency module relies on an attention-driven long short term memory (LSTM) and a time lag matrix to learn its dependencies. Missing values and noise are addressed with total variation reconstruction. Results. We conduct thorough empirical analyses on two real-world datasets. Imputation results show that CGCNImp achieves state-of-the-art performance when compared to previous methods.
关键词	Multivariate time series imputation Graph causal analysis Graph neural network Deep neural network
DOI	10.7717/peerj-cs.966
收录类别	SCI
语种	英语
资助项目	Strategic Priority Research Program of the Chinese Academy of Sciences[XDA19020400] ; National Science Foundation of China[U21B200494]
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence ; Computer Science, Information Systems ; Computer Science, Theory & Methods
WOS记录号	WOS:000795493600002
出版者	PEERJ INC
引用统计	被引频次：1[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/19559
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Liu, Caizheng
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Dept Data Sci, Beijing, Peoples R China 2.Univ Chinese Acad Sci, Dept Comp Sci & Technol, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Liu, Caizheng,Cui, Guangfan,Liu, Shenghua. CGCNImp: a causal graph convolutional network for multivariate time series imputation[J]. PEERJ COMPUTER SCIENCE,2022,8:23.
APA	Liu, Caizheng,Cui, Guangfan,&Liu, Shenghua.(2022).CGCNImp: a causal graph convolutional network for multivariate time series imputation.PEERJ COMPUTER SCIENCE,8,23.
MLA	Liu, Caizheng,et al."CGCNImp: a causal graph convolutional network for multivariate time series imputation".PEERJ COMPUTER SCIENCE 8(2022):23.