A Systematic Literature Review of Spatio-Temporal Graph Neural Network Models for Time Series Forecasting and Classification

A recent article published on arXiv provides a systematic literature review focusing on spatio-temporal graph neural networks (GNNs) and their applications in time series forecasting and classification. The review highlights the increasing interest in GNNs as tools for analyzing dependencies among variables that evolve over time. By offering a comprehensive overview of various modeling approaches, the article sheds light on how these advanced neural network architectures capture complex spatio-temporal relationships. This work contributes to the broader understanding of GNN methodologies within the machine learning community, particularly in contexts where time series data play a critical role. The systematic nature of the review ensures that it covers a wide range of relevant studies, providing valuable insights for researchers and practitioners interested in the intersection of graph-based models and temporal data analysis. Overall, the article serves as a significant resource for advancing knowledge on the use of spatio-temporal GNNs in forecasting and classification tasks.