Graph Neural Network for Spatiotemporal Data: Methods and Applications

• DOI: 10.2139/ssrn.4725185
• Publication Date: 2024-02-13T19:24:31Z
• Authors: Yun Li; Dazhou Yu; Zhenke Liu; Minxing Zhang; Xiaoyun Gong; Liang Zhao
• Keywords: FOS: Computer and information sciences; Computer Science - Machine Learning; Artificial Intelligence (cs.AI); Computer Science - Artificial Intelligence; 11. Sustainability; 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology; Machine Learning (cs.LG)

Abstract: In the era of big data, there has been a surge in availability data containing rich spatial and temporal information, offering valuable insights into dynamic systems processes for applications such as weather forecasting, natural disaster management, intelligent transport systems, precision agriculture. Graph neural networks (GNNs) have emerged powerful tool modeling understanding with dependencies to each other dependencies. There is large amount existing work that focuses on addressing complex spatiotemporal using GNNs. However, strong interdisciplinary nature created numerous GNNs variants specifically designed distinct application domains. Although techniques are generally applicable across various domains, cross-referencing these methods remains essential yet challenging due absence comprehensive literature review data. This article aims provide systematic overview technologies domain. First, ways constructing graphs from summarized help domain experts understand how generate types Then, categorization summary presented enable identify suitable support model developers advancing their research. Moreover, significant offered introduce broader range experts, assisting them exploring potential research topics enhancing impact work. Finally, open challenges future directions discussed.