On Mon, 24 Feb 2003, Sirotkin, Alexander wrote:
> I'm trying to find a way to compare two clusterings (two results of a
> clustering algorithm).
> Is there any algorithm (or better yet - working softwar package for
> S-Plus/R/whatever) ?

What kind of comparison are you looking for?  There are several ways to
compute the agreement of two partitions, but that's probably not what
you're seeking.  If you want to answer that age-old question, "which one
is better?" then you need to have some way to define what "good"  means
(with respect to your clustering task).  There's no generic method that
applies to all problems for answering this question.


--------- Kiri Wagstaff, Ph.D. -------- [log in to unmask] ----------
         They would not find me changed from him they knew --
              Only more sure of all I thought was true.
                     -- Robert Frost, Into My Own