[R-sig-Geo] Spatial Clustering in R

Jim Bouldin jrbouldin at ucdavis.edu
Tue Oct 28 22:49:47 CET 2008

Hi Dan.  The attributes are not distances nor any type of
similarity/dissimilarity measure, but rather actual attributes of
individual spatially referenced items (specifically, continuous and
categorical measures of tree attributes, e.g. diameter and taxon).  Thanks.

And by the way the correct terminology appears to be either spatial
segmentation or spatial regionalization, not the "spatial clustering" that
I used in the subject line of my original post.  I don't mean to refer to
the spatial analysis of the points themselves, but rather to the ATTRIBUTES
of the items at those points.


Hi Jim,

What are the nature of the attributes of the items, are they simply
actually distances from one another, or is their some other measures of
similarity/dissimilarity (say differences in soil types)?

Dan Putler

On Tue, 2008-10-28 at 13:35 -0700, Jim Bouldin wrote:
> I'm wondering whether any R package has the ability to perform spatial
> clustering, i.e. the spatial partitioning of a data set into relatively
> homogeneous polygons, based on one or more attributes of the items (similar
> to what the program eCognition does).  Perhaps there is a more
> statistically correct term for what I have just described, but I don't know
> it.  Thanks.
> Jim Bouldin, PhD
> Associate in Agricultural Experiment Station
> Department of Plant Sciences, UC Davis
> Davis CA, 95616
> 530-554-1740
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Dan Putler
Sauder School of Business
University of British Columbia

More information about the R-sig-Geo mailing list