Authors: Somayeh Dodge*, University of California
Topics: Geographic Information Science and Systems, Qualitative Methods, Temporal GIS
Keywords: cartography, machine learning, movement pattern mining, movement interaction, trajectory visualization
Session Type: Paper
Presentation File: No File Uploaded
Movement is realized in both a three (or four) dimensional space (i.e. location and time) and a multidimensional attribute space (i.e. context variables). The syntheses of these two spaces need new effective tools for dynamic visualization of the traversal of a moving individual through these dimensions. This presentation highlights the importance of geographic visualization as an exploratory analysis tool in the study of movement and interaction among moving entities. I introduce a data science framework for movement in which data-driven analytics and theory-driven models are informed through visualization and domain information to enhance fundamental knowledge of movement. Cartographic theories and visuals principles needed for meaningful representation of motion in space and time and across scales are discussed. Using dynamic visualizations of GPS tracking data enriched with environmental variables, I will show how cartography can support knowledge discovery and machine learning approaches.