involves the use of computer graphics to stimulate the human visual
system to recognize patterns that would not otherwise be obvious. Geovisualization
can be extremely beneficial in all stages of exploratory data analysis
(EDA). Below are some examples of geographic visualization in the Apoala
of a time-space cube in IBM Data Explorer allows a climatologist to
visually search for periodicity in high-intensity rainfall events in
both time and space. In this case, daily total precipitation data over
a 1 year period are displayed for the Susquehanna River Basin in Pennsylvania.
image above is a parallel coordinate plot with certain data records
highlighted using brushing. This brushing technique can be linked to
other representations including maps, scatter plots (see below), and
the data cubes illustrated above in order to allow a user to interactively
explore subsets of the data.
is an example of linked brushing, as described above, in a scatterplot
3-dimensional shaded relief representation of a portion of Pennsylvania
uses color to represent maximum daily temperature. The ability to display
multiple data sets at once, combined with the power to interactively
change the display of such data, helps users quickly and intuitively
explore their data. Visualization can help users to both confirm existing
hypothesis as well as formulate new ones.