Date: Mon, 19 Sep 2005 20:17:24 +0530
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Deepak Jethwani <email@example.com>
Subject: Clustering Zip Codes
Content-Type: text/plain; charset=ISO-8859-1
I am facing the following problem. We have a list of zip codes and a table
that lists out the drive time distance from one zip code to the other in the
Zip1 Zip2 Zip3 .....
Zip1 0 5 15
Zip2 5 0 12
Zip3 15 12 0
Now we need to cluster zip codes which fall within say 10 minutes drive
distance from a zip code into one cluster. The attributes of the cluster
such as number of establishments etc would be aggregated for each cluster.
I have been thinking about the problem and feel that an important aspect is
identifying starting points around which we could form clusters on and how
to prevent overlap of zip codes between these clusters.
Kindly share with me if any one has faced a similar problem or any
suggested approaches that come to mind for a possible solution.