Re: finding the centre of a cluster
- From: "Nick" <tulse04-news1@xxxxxxxxxxx>
- Date: Mon, 4 Jun 2007 00:00:54 +0100
"Reef Fish" <large_nassua_grouper@xxxxxxxxx> wrote in message
news:1180906471.022903.89400@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
On Jun 3, 8:14 am, "Nick" <tulse04-ne...@xxxxxxxxxxx> wrote:
"Question" <breceivem...@xxxxxxxxx> wrote in message
news:1180778044.224598.126320@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Is there any mathematical way for finding centre of clusters in
clustering issue?
As per Phil Sherrod, see
http://www2.chass.ncsu.edu/garson/PA765/cluster.htm
and look up "Cluster centers".
"Cluster centers are the average value on all clustering variables of
each
cluster's members"
That's an arbitrary definition of the "center" of a cluster.
The "centroid method" is one way to find the "center" of a
cluster. There are other criteria just as arbitrary. Why
would anyone want to know the center of a cluster anyway?
I didn't ask the question. But it is a long time ago, but something to do
with variability within clusters as between clusters. To do that AFAIR one
would have to have some measure of variability, which is generally based on
distance from x bar.
I might add that it is nearly 30 years since I have been working with this
material.
Nick
.
- Follow-Ups:
- Re: finding the centre of a cluster
- From: Reef Fish
- Re: finding the centre of a cluster
- References:
- finding the centre of a cluster
- From: Question
- Re: finding the centre of a cluster
- From: Nick
- Re: finding the centre of a cluster
- From: Reef Fish
- finding the centre of a cluster
- Prev by Date: Re: finding the centre of a cluster
- Next by Date: Re: finding the centre of a cluster
- Previous by thread: Re: finding the centre of a cluster
- Next by thread: Re: finding the centre of a cluster
- Index(es):
Relevant Pages
|