Contextualizing human dynamics: Understanding the semantics of movement trajectories with Wi-Fi data

Abstract

The increasing popularity of Wi-Fi enabled smart devices, such as cell phones, poses unprecedented opportunities for modeling large-scale human movement trajectories. While these datasets enable the extended spatiotemporal scales and refined granularities for human dynamics research, the lack of semantic information hinders the understanding of the processes and rationales behind human movements. In this study, we employ Wi-Fi detectors to identify track points along travelers’ movement trajectories. Also, we propose a semantic space–time matrix (SSTM) method to derive the semantic information from the movement trajectories. The temporalities and activity types of track points are aggregated to understand the mobility patterns of different travelers. We have further applied the method to a small tourist community, the Shichahai scenic area of Beijing, China, with the Wi-Fi data collected from 463,874 individuals over two weeks. The derived semantic information has been further validated by a field survey, justifying the feasibility of Wi-Fi data for movement tracking and semantic analysis. The Wi-Fi data and the proposed analytical method can shed new insights into human dynamics research by enriching the context of individual travel.