Spatio-Temporal Analysis of Aggregated Human Activities Based on Massive Mobile Phone Tracking Data

Abstract

Urban space and the behavior of human activities constantly interact with each other. Investigation on distribution of aggregated human activities and spatio-temporal change benefits data-driven policy-making in urban planning and urban governing. In the era of big data, with the development of information and communication technologies, it is possible to collect city-scale data with high resolution in space and time by various location-aware devices and sensors. Exploration of spatial-temporal activities attracts a lot of attention. By taking about 10 million one-day tracking data of mobile phone users in Shenzhen, China as an example, this paper firstly identified their stay locations according to spatial and temporal rules to generate stay trajectory for each individual and recovered activity semantic information by labelling activity types for each stay locations. Then, the significant differences in patterns of distributions of stay locations and their activities were analyzed. Spatial and temporal distributions of different human activities were explored, respectively. The study shows that the distribution of stay locations and activities is obviously heterogeneous. The average number of stay locations of an individual per day is 2.1, while the average number of activities an individual engaged in per day is 3.4. This study furthermore suggests that different types of activities have temporal variance and spatial heterogeneity. The temporal distribution fluctuates significantly over 24 hours, which is in accordance with daily routine. The spatial distribution overall obeys ``space power law'', and the spatial distribution of social activity, which has a faster-down tail, shows a more obvious pattern of spatial segregation than the other two activities. The study revealed the diversity and heterogeneity of spatial and temporal distribution of human aggregated activities in urban space, which is meaningful in analyzing human activities research and facilitating urban traffic optimization and urban planning.

Publication
Journal of Geo-Information Science
Jinzhou Cao(曹劲舟)
Jinzhou Cao(曹劲舟)
Assistant Professor

My research interests Urban big data mining, Geo-AI and Urban Analytics.

Rui Cao
Rui Cao
Undergraduate Student