Graph spatiotemporal process for multivariate time series anomaly detection with missing values

Yu Zheng, Huan Yee Koh, Ming Jin, Lianhua Chi, Haishuai Wang, Khoa Tran Phan, Yi-Ping Phoebe Chen, Shirui Pan, Wei Xiang

Abstract: Highlights•This paper presents a novel anomaly detection method for multivariate time series with missing values.•The method employs graph neural networks to capture both spatial and temporal dependencies.•A distribution-based anomaly scoring mechanism is used to detect anomalies.