Understanding human movement dynamics from different big movement data – a comparative analysis

Authors: Junchuan Fan*, CGIS, University of Maryland, Kathleen Stewart, Center for Geospatial Information Science, University of Maryland
Topics: Geographic Information Science and Systems, Urban Geography, Transportation Geography
Keywords: Human mobility, big data, social media, GPS, cellphone data
Session Type: Paper
Day: 4/4/2019
Start / End Time: 8:00 AM / 9:40 AM
Room: Congressional B, Omni, West
Presentation File: No File Uploaded


The ubiquitity of location-aware sensors as well as increasing data storage and processing capacity make it possible to collect massive digital traces of human activities and study underlying human movement dynamics at different spatiotemporal granularities. Researchers have investigated human movement dynamics at individual as well as aggregate levels using different kinds of movement data, e.g., cellphone records, GPS vehicle trajectory data, subway smart-card records, and geo-tagged social media data. Most of these previous studies, however, rely on data from one kind of device/sensor, and thus lack a robust understanding of the unique characteristics of movement data collected from different sources, which can be significant should findings derived from these movement data be used to inform policy decision-making. In this research, using four kinds of movement data, namely, cellphone records, location-based mobile app data, vehicle GPS trajectory data and geotagged social media data collected during July 2017 for Baltimore-Washington D.C. area, we analyzed and compared the same human mobility measures derived from different big movement data. To investigate how different spatiotemporal sampling strategies of different movement data capture mobility differently, we analyzed four mobility measures including distribution of displacement distance, inter-activity time, radius of gyration, location exploration probability at the individual level, and compared the results with previous findings in the literature; at the aggregate level, we analyzed and compared the temporal distribution of activities at the same place for different types of places (e.g., residential area, shopping center).

Abstract Information

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

To access contact information login