A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection

TL;DR

This survey addresses the challenge of applying graph neural networks to time series data (GNN4TS) by organizing the field into four core tasks—forecasting, classification, anomaly detection, and imputation—and presenting a unified spatial-temporal framework. It analyzes how spatial (inter-variable) and temporal (inter-temporal) dependencies are modeled through spectral, spatial, and hybrid GNNs, and how these modules are architecturally fused in discrete or continuous forms. The paper comprehensively reviews representative methods, datasets, and practical applications across domains such as transportation, energy, and healthcare, and outlines future directions including pre-training, robustness, privacy, scalability, and AutoML. Overall, it provides a detailed, actionable roadmap for researchers and practitioners seeking to design, implement, and evaluate GNN-based time series models across diverse applications.

Abstract

Time series are the primary data type used to record dynamic system measurements and generated in great volume by both physical sensors and online processes (virtual sensors). Time series analytics is therefore crucial to unlocking the wealth of information implicit in available data. With the recent advancements in graph neural networks (GNNs), there has been a surge in GNN-based approaches for time series analysis. These approaches can explicitly model inter-temporal and inter-variable relationships, which traditional and other deep neural network-based methods struggle to do. In this survey, we provide a comprehensive review of graph neural networks for time series analysis (GNN4TS), encompassing four fundamental dimensions: forecasting, classification, anomaly detection, and imputation. Our aim is to guide designers and practitioners to understand, build applications, and advance research of GNN4TS. At first, we provide a comprehensive task-oriented taxonomy of GNN4TS. Then, we present and discuss representative research works and introduce mainstream applications of GNN4TS. A comprehensive discussion of potential future research directions completes the survey. This survey, for the first time, brings together a vast array of knowledge on GNN-based time series research, highlighting foundations, practical applications, and opportunities of graph neural networks for time series analysis.

Figures (6)

• Figure 1: Graph neural networks for time series analysis (GNN4TS). In this example of a wind farm, different analytical tasks can be categorized into time series forecasting, classification, anomaly detection, and imputation.
• Figure 2: Examples of spatial-temporal graphs, where node colors represent distinct features. The top and bottom panels demonstrate spatio-temporal graphs with fixed and dynamic graph structures over time, respectively.
• Figure 3: Task-oriented taxonomy of graph neural networks for time series analysis in the existing literature.
• Figure 4: Four categories of graph neural networks for time series analysis. For the sake of simplicity and illustrative purposes, we assume the graph structures are fixed in all subplots.
• Figure 5: Methodology-oriented taxonomy of graph neural networks for time series analysis.

Theorems & Definitions (5)

• Definition 1: Univariate Time Series
• Definition 2: Multivariate Time Series
• Definition 3: Attributed Graph
• Definition 4: Spatial-temporal Graph
• Definition 5: Graph Neural Network