Trajectory Understanding and Spatial-temporal Profile Learning Based on Mobile Crowdsourced Data

Authors: Chih-Yuan Chen*, Department of Geography, Chinese Culture University, Hao-Chun Hsu , Department of Geography, Chinese Culture University, You-Ming Xu, Department of Geography, Chinese Culture University, Pei-Yu Luh, Department of Geography, Chinese Culture University, You-Ru Lin, Department of Geography, Chinese Culture University, Chih-Chi Kao, Department of Geography, Chinese Culture University
Topics: Geographic Information Science and Systems, Spatial Analysis & Modeling
Keywords: mobile crowdsourced data, profile learning, trajectory, geographic knowledge discovery
Session Type: Poster
Day: 4/5/2019
Start / End Time: 9:55 AM / 11:35 AM
Room: Lincoln 2, Marriott, Exhibition Level
Presentation File: No File Uploaded


Crowdsoured data from mobile devices such as smartphones and internet of things gadgets can provide meaningful information about users' spatio-temporal context. By analyzing users locations and trajectories, service providers can extract meaningful information and hidden patterns from the data and thereafter provide better context-aware services. In this project, we devided spatial-temporal profiling of mobile crowdsoured data into two categories, fast dynamics and slow dynamics. The fast-dynamic profile includes the user trajectories, transportation patterns, and the other environmental factors such as nearest POIs while the slow-dynamic profile contains the user age, sex, interests, and the other attributes based on predicted and pre-collected data. The “Divide and Group” strategy is the main idea for our research to find the knowledge for the fast-dynamic profile and slow-dynamic profile. We used this strategy on real smartphones' sensory data and concluded several spatio-temporal contexts of designated user groups.


Abstract Information

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

To access contact information login