Graph Neural Networks for Spatiotemporal Data Analysis in Smart Cities

Abstract

Smart cities generate vast amounts of spatiotemporal data from various sources such as sensors, social media, and transportation systems. In this paper, we investigate the application of graph neural networks (GNNs) for analyzing and extracting insights from complex spatiotemporal data in smart city environments. Our research explores different GNN architectures and demonstrates their efficacy in tasks such as traffic prediction, anomaly detection, and urban planning.