Databases Reference
In-Depth Information
You will be prompted for a name and description. At a minimum, give it a
name.
Select “Create”.
The new dataset now appears on the Control Center - derived from the Pollen
dataset.
View the summary statistics for this subset. Notice that, using the PCP, we
have accomplished both observation reduction (almost 4,000 observations
down to about 100) and dimension reduction (ObsNo was removed from
the plot).
Close the summary statistics.
View the derived set in a scatter plot.
What do you see? Has your data mining activity found any patterns of
interest?
There are numerous other options in the PCP that have not yet been described.
You are encouraged to experiment with them on your own. Many of the options
become available with a right-click of either a range slider for filter options, or an
axis. All features of the filter option are also available using the icon buttons in
the options panel on each filter bar, except for the “Make Dataset ...”option.
You have already used the “Hide” option of an axis to remove it from the plot.
The “Add Marker” option assists in point value reading. As noted earlier, the
axes were designed without scales in order to reduce chart clutter and to get
the user to focus more on patterns than value reading. If you need to know the
value at a given location on an axis, then hover over that location with the mouse
pointer and the value at the hover location will be displayed. If you would
like the value to remain after moving the mouse pointer away:
Right-click at the desired location on the axis.
Select the “Add Marker” option.
Move to a different location along an axis, then add another marker.
Hover over an existing marker and right-click to remove the marker or
alternatively . . .
Select the delete “Markers” button in the options panel to remove all
displayed markers.
When plotting density heights on the 3-D Z axis, VisMiner uses the point
of greatest density over all axes as the top point. That top point is always
