Pitney Bowes MapInfo Vertical Mapper - Manual (Page 184)

Point Aggregation with Statistics
182
Vertical Mapper 3.7
Original distribution of data points.
The Cluster Density Aggregation technique is usually most effective for small to medium size data
sets, because it makes better decisions during the aggregation process. It aggregates the more
densely clustered points first within a reasonable processing time. However, it is not appropriate if
you need to know the number of points aggregated or if there are coincident points in the data.
Coincident points are always aggregated first and are not included in the statistics appended to the
new point file. The alternative is to use the Forward Stepping Aggregation technique.
The Cluster Density Aggregation technique looks at the entire data set prior to aggregation and
determines the single most densely populated area that would fall inside the user-specified search
radius (aggregation distance). Points that fall inside the search radius are chosen and flagged. The
geocenter of these points is calculated and that position becomes the location of the new
aggregated point. Calculations are performed on the values of the selected points (as specified in
the Point Aggregation dialog box) and the results are attributed to the new geocentred point. Then,
the area with the second highest density of points is chosen and the process is repeated. At each
stage, the entire remaining data set must be examined for its density patterns to avoid using
previously aggregated points and to factor in the removed points in the density analysis.
Cluster density aggregation always includes coincident point handling regardless of whether or not
you request this type of handling, and the point density calculation used in the aggregation
technique does not handle points that are coincident. If you do not choose any of the settings on the
Coincident Point Handling dialog box, the default values will be used. A warning message appears
reminding you when this technique is chosen.
An example of circle aggregation using the Cluster Density Aggregation technique.