BookmarkSubscribeRSS Feed
mtambors
Calcite | Level 5

I am attempting to run a cluster analysis using the HP Cluster node to identify groups of schools with similar demographic properties. When I run the analysis, I get 12 clusters (Screenshot 1). 

 

 However, when I run the segment profile node (process flow in screenshot 2), all observations get assigned to a single cluster (screenshot 3).

 

I cannot for the life of me understand why the 12 clusters are not being brought over into the segment profile. This only occurs after the HP Cluster run. When I attach the segment profile node to a regular cluster analysis node, it works as expected. Screenshot 4 shows the properties of the segment profile node.

 

Any help would be greatly appreciated.

 

Thanks!

Mike

6 REPLIES 6
jwexler
SAS Employee

Hi, can you send over the properties panel screenshot for your HP Cluster node? I tried your flow using the sample HMEQ dataset that is included with EM, and segment profile worked as expected. I also wonder if your segment sizes are too small for segment profile, because there may be a cutoff value.

 

Thanks

Jonathan

 

hp cluster.PNG

mtambors
Calcite | Level 5
[cid:c011d60b-73fa-4902-bab1-95d3e95bb506]

I actually heard this morning that the problem may be due to the fact that I only have the desktop version of EM, which the HP nodes don't work well on. Could that be the reason?


The analysis itself returned 14 clusters with frequencies ranging from 32 to 374.


Thanks!

Mike
jwexler
SAS Employee

That should not matter for what you are doing. Can you try creating your HP Cluster with a max of 2 clusters?  Lets see if your problem is data-dependent.

mtambors
Calcite | Level 5
I'm still getting the same result.


The results of the HP cluster is showing 2 clusters:


[cid:8b6c10a8-f60b-4bb2-b510-95a0262c50eb]


However, when I explore the exported data set, every cluster ID is 1:


[cid:05cf8056-0edf-456e-80f4-72c39b8bdb7f]


Oddly enough, I was able to get it to work using the HMEQ dataset as well, so I am not sure what the difference is.

jwexler
SAS Employee

That's great that it's working now. It looks like it's data dependent, as I suspected. Somehow your obersvations are being scored/put into 1 cluster.

 

Best advice for a quick solution is to open up a quick tech support track to get hands-on advice from our trained staff.

 

support.sas.com

 

 

Good luck!

SAS Innovate 2025: Save the Date

 SAS Innovate 2025 is scheduled for May 6-9 in Orlando, FL. Sign up to be first to learn about the agenda and registration!

Save the date!

How to choose a machine learning algorithm

Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 6 replies
  • 2594 views
  • 0 likes
  • 2 in conversation