Using GPS to learn significant locations and predict movement across multiple users
Personal and Ubiquitous Computing
From trajectories to activities: a spatio-temporal join approach
Proceedings of the 2009 International Workshop on Location Based Social Networks
Location-based activity recognition using relational Markov networks
IJCAI'05 Proceedings of the 19th international joint conference on Artificial intelligence
Data-driven trajectory smoothing
Proceedings of the 19th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
Inferring human activities from GPS tracks
Proceedings of the 2nd ACM SIGKDD International Workshop on Urban Computing
| Hi-index | 0.00 |
GPS (Globe Positioning System) trajectory data provide a new way for city travel analysis others than traditional travel diary data. But generally raw GPS traces do not include information on trip purposes or activities. Earlier studies addressed this issue through a combination of manual and computer-assisted data processing steps. Nevertheless, geographic context databases provide the possibility for automatic activity identification based on GPS trajectories since each activity is uniquely defined by a set of features such as location and duration. Distinguished with most existing methods using two dimensional factors, this paper presents a novel approach using spatial temporal attractiveness of POIs (Point of Interests) to identify activity-locations as well as durations from raw GPS trajectory. We also introduce an algorithm to figure out how the intersections of trajectories and spatial-temporal attractiveness prisms indicate the potential possibilities for activities. Finally, Experiments using real world GPS tracking data, road networks and POIs are conducted for evaluations of the proposed approach.