Root-cause analysis for time-series anomalies via spatiotemporal graphical modeling in distributed complex systems

作者：

Highlights：

• Framing root-cause analysis as a minimization problem using inference based metric.

• Node inference algorithm to isolate the possible failed node (variable or subsystem).

• Validation in terms of accuracy, scalability, robustness, adaptiveness and efficiency.

• The proposed approach outperforms state-of-the-art methods on synthetic and real data sets.

摘要

•Framing root-cause analysis as a minimization problem using inference based metric.•Node inference algorithm to isolate the possible failed node (variable or subsystem).•Validation in terms of accuracy, scalability, robustness, adaptiveness and efficiency.•The proposed approach outperforms state-of-the-art methods on synthetic and real data sets.

论文关键词：Distributed complex system,Anomaly detection,Root cause analysis

论文评审过程：Received 13 June 2020, Revised 23 August 2020, Accepted 12 October 2020, Available online 19 October 2020, Version of Record 20 October 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106527