Investigating intra-urban human mobility patterns based on aggregate mobile phone data: A Bayesian spatio-temporal analyses

Authors: Zhensheng Wang*, Shenzhen Key Laboratory of Spatial Smart Sensing and Services, Shenzhen University, Shenzhen, 518060, Yang Yue, Shenzhen Key Laboratory of Spatial Smart Sensing and Services, Shenzhen University, Shenzhen, 518060, Qingquan Li, Shenzhen Key Laboratory of Spatial Smart Sensing and Services, Shenzhen University
Topics: Spatial Analysis & Modeling, Urban Geography, Applied Geography
Keywords: Urban dynamics, Bayesian hierarchical models, space-time interactions, human mobility patterns
Session Type: Paper
Day: 4/11/2018
Start / End Time: 8:00 AM / 9:40 AM
Room: Oakley, Sheraton, 4th Floor
Presentation File: No File Uploaded


The remarkable development of the Information and Communication Technologies (ICTs) has changed every single aspect of our lives, emotionally and mentally. Through the application of Bayesian spatio-temporal analysis to aggregate mobile phone data of Shenzhen Municipality, China, the primary goal of this research is to analyze urban dynamics of Shenzhen, China, based on the numbers of mobile phone users in a 24-hour period and discuss a class of Bayesian hierarchical models that simultaneously characterize spatial and temporal patterns. Specifically, our purpose is a) to explore the overall relationship between aggregate mobile phone users and time at a small-area level and b) to identify the geographic variation of area-specific trends and whether there is a spatio-temporal interaction effects and c) to pinpoint “unstable” areas showing evidence of unusual variability in the time pattern of human mobility, such as hot spots and cold spots in the mobile phone data. The results may helpful for the planning and provision of municipal facilities and services. Further time-varying or space-varying coefficients could be considered in the future, allowing for different effects of covariates in time or space.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login