SparseTrajAnalytics: an Interactive Visual Analytics System for Sparse Trajectory Data

• Published in: Journal of geovisualization and spatial analysis Vol. 5; no. 1
• Main Authors: Ye, Xinyue, Du, Jiaxin, Gong, Xi, Zhao, Ye, AL-Dohuki, Shamal, Kamw, Farah
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01-06-2021
• Subjects: cities; Earth and Environmental Science; Geographical Information Systems/Cartography; Geography; Geology; Landscape/Regional and Urban Planning; Remote Sensing/Photogrammetry; towns; Urban Geography; Urbanism (inc. megacities; Hurricane trajectories; Visual analytics; Sparse trajectories; Human migration
• ISSN: 2509-8810; 2509-8829
• DOI: 10.1007/s41651-020-00068-1

Sparse trajectory data are trajectories that cover a relatively large geographic area with infrequent samplings of movements, such as human migration and hurricane trajectories. Compared with dense trajectory data (e.g., massive vehicle movements in a metropolitan area), sparse trajectories represent a more general form of movement data due to the data availability and less sensitivity. However, there is a lack of open source software for sparse trajectory data analytics. We designed an online system with a self-adaptive heatmap and multi-spatial-temporal-view charts to analyze sparse trajectories both visually and interactively. With our system, users can (1) store and manage sparse trajectory data on a cloud service; (2) analyze sparse movement behaviors by querying consolidated and live data sets using keywords, spatial location, and time constraints; and (3) explore query results and associated data through web-based thematic maps, tables, and charts. Results of the survey among domain experts show that the open source system is intuitive and user-friendly.