Re: Fuzzy clustering

Xavier Otazu (xotazu@icc.es)
Wed, 3 Nov 1999 12:37:11 +0100 (MET)

PPC wrote in message <7vonds$dsf$1@nslave2.tin.it>...
>The problem: I have thousands of points in a 2D space characterized by
their
>position (value,possibility) and by a multiplicity
>I display them in a bubble graph with each bubble's radius proportional to
>multiplicity
>These points concentrate in 2 or more zones (graphically)
>I'd like to get information about these zones through data analysis
>Does anybody have a clustering algorithm that can fit this issue?
>

From your explanation I don't think you necessarily need to use any
fuzzy tool.

You could try the usual ISODATA algorithm for cluster analysis. If you
need to use your 'multiplicity' value, you can use it in the ISODATA
algorithm as the mass of each point to get the mass center for each cluster.
This way each point will have a different influence when searching the
cluster.

If you don't know the ISODATA algorithm, don't worry. It is very easy to
understand and to implement. You can find this algorithm on almost any book
on image classification.

############################################################################
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
fuzzy-owner@dbai.tuwien.ac.at
(4) WWW access and other information on Fuzzy Sets and Logic see
http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html