Spatio-Temporal Graph Attention Embedding for Joint Crowd Flow and Transition Predictions: A Wi-Fi-based Mobility Case Study

Abstract

Crowd mobility prediction, in particular, forcasting flows at and transitions across different locations, is essential for crowd analytics and management in spacious environments featured with large gathering We propose GAEFT, a novel rowd moility analytis system based on the multi-task graph attention neural network to forecast rowd flow Extensie experimental studies using more than 28 million associateion records collected during 2020-2021 academic year validate the excellent accuracy of GAEFT in forecastin dynamic and comple crowd mobility.