a problem interpreting hierarchical cluster output

a problem interpreting hierarchical cluster output

Post by Patrick van Lonkhuijze » Fri, 04 Feb 2000 04:00:00



Hello people,

I have a little difficulty interpreting the spss output of a hierarchical
cluster
analysis what does the agglomerate schedule of the average linkage represent
when a hierarchical cluster analysis has been done?

 
 
 

a problem interpreting hierarchical cluster output

Post by Rich Ulric » Fri, 04 Feb 2000 04:00:00


On Thu, 3 Feb 2000 12:36:03 +0100, "Patrick van Lonkhuijzen"


> I have a little difficulty interpreting the spss output of a hierarchical
> cluster
> analysis what does the agglomerate schedule of the average linkage represent
> when a hierarchical cluster analysis has been done?

order, isn't it?
--

http://www.pitt.edu/~wpilib/index.html

 
 
 

a problem interpreting hierarchical cluster output

Post by Ulrike Tim » Thu, 10 Feb 2000 04:00:00


The agglomeration schedule shows every step of the cluster agglomeration. if you
have n cases, at first step, it creates n-1 clusters, at second step n-2
clusters and so on until the last step, where it has agglomerated all cases into
one cluster. in the column in the middle the distance coefficient is shown. it
slowly increases from step to step. you should stop clustering at the one step
when the distance coefficient becomes immediately higher. (to find out, how many
clusters you have then, count from the last step backwards!)

good luck,
ulrike


> Hello people,

> I have a little difficulty interpreting the spss output of a hierarchical
> cluster
> analysis what does the agglomerate schedule of the average linkage represent
> when a hierarchical cluster analysis has been done?

 
 
 

a problem interpreting hierarchical cluster output

Post by Patrick van Lonkhuijze » Sat, 12 Feb 2000 04:00:00


thanks, I found I have a little problem because of the fact that I'm using
14134 observations when I tried with only 10, to gain a little insight, I
noticed how it worked I couldn't find it out because I had to many
observations which made it rather difficult to interpret when one doesn't
know how to. but it is good to know I figured it out the right way.

so thanks

Patrick


>The agglomeration schedule shows every step of the cluster agglomeration.
if you
>have n cases, at first step, it creates n-1 clusters, at second step n-2
>clusters and so on until the last step, where it has agglomerated all cases
into
>one cluster. in the column in the middle the distance coefficient is shown.
it
>slowly increases from step to step. you should stop clustering at the one
step
>when the distance coefficient becomes immediately higher. (to find out, how
many
>clusters you have then, count from the last step backwards!)

>good luck,
>ulrike


>> Hello people,

>> I have a little difficulty interpreting the spss output of a hierarchical
>> cluster
>> analysis what does the agglomerate schedule of the average linkage
represent
>> when a hierarchical cluster analysis has been done?

 
 
 

1. deriving intial cluster seeds from hierarchical clustering

Dear Friends:

I am conducting a cluster analysis and am following
the traditional method of conducting a hierarchical
cluster on a sample size first to generate a set of
feasible solutions. Then, I will feed each initial
solution into a k-mean clustering (proc fastclus) to
generate clusters. The final solution will be then
determined by some internal and external validation,
along with business inputs.

As you know, proc cluster (hierarchical) generate a
set of solutions all the from from one to say 40
clusters. I then look at the cubic clustering
criterion and select several feasible solutions.
Suppose I may want to try cluster # =4,5,6 all in
using k-mean. What I also like to do is to extract the
corresponding cluster centeriods to be used as the
initial seeds for the k-means. Can someone help me
with this problem? I checked the SAS manual and it
does not see to have an example.

thanks.

Hongjie

__________________________________________________
Do you Yahoo!?
U2 on LAUNCH - Exclusive greatest hits videos
http://launch.yahoo.com/u2

2. Run-time linking RASAP32.DLL problems...

3. Interpreting PROC NLIN output?

4. Which SDK APIs to use for Network Run Count checking?

5. interpreting time series output

6. CD file linking program?

7. How to interpret ASEs in PROC FREQ output

8. OO-Course taught with C++

9. How to interpret Monte Carlo Simulation in SPSS outputs

10. Interpreting GENLOG output

11. Problem: SAS (SUN) per ODBC: Integer are interpreted as $6.1

12. Releasing SCore Cluster System Software on PC cluster

13. Combining clusters of size 1 w/bigger clusters?