基于网络和图的时空智能——概念、方法和应用

当前时空智能(SpaceTimeAI)和地理空间智能(GeoAI)已是热门的话题,该研究领域旨在将计算机科学的最新方法(如深度学习)应用于地理空间问题。虽然深度学习方法因其对栅格数据的自然适用性而在图像处理中取得了巨大成功,但仍未广泛应用于其他空间和时空数据类型。本文提出使用网络和图作为SpaceTimeAI或GeoAI的基本结构的倡议,并将其应用于城市研究中。相比于基于网格的表达,基于网络的结构更加精确和实用。图能实现对点、线、面/多边形/网格和网络等多种空间结构的表达。本文通过时空预测、聚类和时空优化等常用时空分析方法展示基于网络和图的时空智能分析的优势,并介绍其在交通出行、警务和公共卫生等领域的应用。

Network and graph-based SpaceTimeAI: conception,method and applications

SpaceTimeAI and GeoAI are currently a hot topic, which apply the latest algorithms in computer science, such as deep learning. Although deep learning algorithms have been successfully applied in raster data processing due to their natural applicability to image processing, their applications in other spatial and space-time data types are still immature. This paper sets up the proposition of using the network (& graph)-based framework as a generic spatial structure to present space-time processes that are usually represented by the point, polyline, and polygon. We illustrate network and graph-based SpaceTimeAI, from graph-based deep learning for prediction, to clustering and optimisation. These demonstrate the advantages of the network (graph)-based SpaceTimeAI in the applications of transport & mobility, crime & policing, and public health.