Predicting Citywide Crowd Flows in Irregular Regions Using Multi-View Graph Convolutional Networks

Junkai Sun, Junbo Zhang, Qiaofei Li, Xiuwen Yi, Yuxuan Liang, Yu Zheng

Abstract: Being able to predict the crowd flows in each and every part of a city, especially in <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">irregular regions</i> , is strategically important for traffic control, risk assessment, and public safety. However, it is very challenging because of interactions and spatial correlations between different regions. In addition, it is affected by many factors: i) multiple <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">temporal correlations</i> among different time intervals: closeness, period, trend; ii) complex <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">external</i> influential factors: weather, events; iii) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">meta</i> features: time of the day, day of the week, and so on. In this paper, we formulate crowd flow forecasting in irregular regions as a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">spatio-temporal graph</i> (STG) prediction problem in which each node represents a region with time-varying flows. By extending <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">graph convolution</i> to handle the spatial information, we propose using <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">spatial graph convolution</i> to build a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">multi-view graph convolutional network</i> (MVGCN) for the crowd flow forecasting problem, where different views can capture different factors as mentioned above. We evaluate MVGCN using four real-world datasets (taxicabs and bikes) and extensive experimental results show that our approach outperforms the adaptations of state-of-the-art methods. And we have developed a crowd flow forecasting system for irregular regions that can now be used internally.