fuzzy clusters

Dave Lorenz (lorenz@usgs.gov)
Sat, 31 Jan 1998 16:52:08 +0100 (MET)

I recently analyzed some data using a fuzzy cluster technique. Some of the
membership scores were counter-intuitive, at least to what I think about fuzzy
membership. I noticed that those points that were very close to a cluster
center were given low membership scores to other clusters, but points that
were near the hull (but close to that cluster were given higher memberships in
other clusters that were farther away than in the first case (see diagram).

|
|
| .02
| + X
| .08
|
|____________________________________

In this diagram, there are two cluster centers, marked by a + and an X, and
the points in question are represented by .02 and .08. My expectation is that
the point marked by .02 membership in X should actually have a larger
membership score that it now has because it is much closer to X than is the
point that has a .08 score. If the clusters are in fact fuzzy, then why
should proximity to a cluster center be given so much weight?
Comments? Thanks.

-- 
Dave Lorenz (lorenz@usgs.gov)
U.S. Geological Survey
2280 Woodale Drive
Mounds View, MN 55112

-- 
Dave Lorenz (lorenz@usgs.gov)
U.S. Geological Survey
2280 Woodale Drive
Mounds View, MN 55112