Mobility Graph

Mobility graphs represent human movement patterns as networks, aiming to analyze and predict various phenomena influenced by these patterns, such as traffic flow, disease spread, and location estimation. Current research heavily utilizes graph neural networks (GNNs), often incorporating spatio-temporal information and fusing data from multiple sources like telecom records, WiFi signals, and street view imagery, to improve model accuracy and robustness. These advancements enable more accurate predictions and improved understanding of complex urban systems, with applications ranging from optimizing urban planning and transportation to enhancing public health surveillance.