Graph Anomaly Detection in Time Series: A Survey

2023-01-31

AI Summary

Time-series anomaly detection requires joint modeling of intra-variable temporal dependencies and inter-variable structural dependencies—a challenge inadequately addressed by conventional methods. This paper introduces the first unified taxonomy—Graph Neural Network-driven Time-Series Anomaly Detection (G-TSAD)—systematically surveying over 50 state-of-the-art models and revealing the fundamental advantages of graph representations in multivariate time-series modeling. Methodologically, it innovatively integrates dynamic graph construction, spatio-temporal graph convolution (ST-GCN), graph attention networks (GAT), and self-supervised graph learning, rigorously analyzing their respective applicability and limitations. The study identifies scalability, interpretability, and low-label dependency as three critical open challenges, and proposes concrete cross-domain deployment pathways. By synthesizing theoretical insights with practical considerations, this survey provides a systematic foundation for advancing both the theory and real-world application of G-TSAD.

Abstract

With the recent advances in technology, a wide range of systems continue to collect a large amount of data over time and thus generate time series. Time-Series Anomaly Detection (TSAD) is an important task in various time-series applications such as e-commerce, cybersecurity, vehicle maintenance, and healthcare monitoring. However, this task is very challenging as it requires considering both the intra-variable dependency (relationships within a variable over time) and the inter-variable dependency (relationships between multiple variables) existing in time-series data. Recent graph-based approaches have made impressive progress in tackling the challenges of this field. In this survey, we conduct a comprehensive and up-to-date review of TSAD using graphs, referred to as G-TSAD. First, we explore the significant potential of graph representation for time-series data and and its contributions to facilitating anomaly detection. Then, we review state-of-the-art graph anomaly detection techniques, mostly leveraging deep learning architectures, in the context of time series. For each method, we discuss its strengths, limitations, and the specific applications where it excels. Finally, we address both the technical and application challenges currently facing the field, and suggest potential future directions for advancing research and improving practical outcomes.

Problem

• Detect anomalies in time-series data using graphs
• Address intra- and inter-variable dependencies in TSAD
• Review graph-based deep learning techniques for TSAD

Innovation

• Graph representation for time-series anomaly detection
• Deep learning architectures for graph anomaly detection
• Comprehensive review of G-TSAD techniques and applications