Database Reference
In-Depth Information
considering space and time as dimensions of a unified continuum (space-time
cube) and representation of behaviors of individuals as paths in this continuum;
information visualization, with its techniques for user-display interaction sup-
porting exploratory data analysis; and geovisualization, with its interactive maps
and associated methods enabling exploration of spatial information.
This chapter gives a glimpse of the variety of the existing visual analytics
methods for analyzing movement data. We group the methods into four cate-
gories according to the analysis focus:
1. Looking at trajectories: The focus is on trajectories of moving objects consid-
ered as wholes. The methods support exploration of the spatial and temporal
properties of individual trajectories and comparison of several or multiple
trajectories.
2. Looking inside trajectories: The focus is on variation of movement character-
istics along trajectories. Trajectories are considered at the level of segments
and points. The methods support detecting and locating segments with par-
ticular movement characteristics and sequences of segments representing
particular local patterns of individual movement.
3. Bird's-eye view on movement: The focus is on the distribution of multiple
movements in space and time. Individual movements are not of interest;
generalization and aggregation are used to uncover overall spatio-temporal
patterns.
4. Investigating movement in context: The focus is on relations and interactions
between moving objects and the environment (context) in which they move,
including various kinds of spatial, temporal, and spatio-temporal objects and
phenomena. Movement data are analyzed together with other data describing
the context. Computational techniques are used to detect occurrences of
specific kinds of relations or interactions and visual methods support overall
and detailed exploration of these occurrences.
We demonstrate the capabilities of visual analytics by examples using a data
set consisting of GPS tracks of 17,241 cars collected during one week in Milan,
Italy. The data were provided by Comune di Milano (Municipality of Milan).
8.2 Looking at Trajectories
In this section, we consider, first, the techniques for visual representation of
trajectories and interaction with the representations; second, the use of clustering
methods for comparative studies of multiple trajectories; and, third, the time
transformations supporting exploration of temporal properties of trajectories
and comparison of dynamic properties of multiple trajectories.
