Building models with SAS Enterprise Miner, SAS Factory Miner, SAS Visual Data Mining and Machine Learning or just with programming

decision stump and conditional decision trees

Reply
Contributor
Posts: 47

decision stump and conditional decision trees

Hi,

I am using decision trees and trying to find out how can I use decision stump and conditional decision tree and M5.

Is that part of normal DT or part of HP tree etc or I just need to set some important parameters about them?

 

 

Regards

 

SAS Employee
Posts: 37

Re: decision stump and conditional decision trees

Posted in reply to geniusgenie

Hello, GeniusGenie.

 

A 'stump' is simply a tree with one split.  So, set MAXDEPTH=1 in PROC ARBOR or PROC HPFOREST to do that, or the equivalent depth option in HPSPLIT. 

 

Assuming "Conditional Decision Trees" refers to the ideas in "conditional inference trees" (Hothorn, Hornik, and Zeileis 2006), then use PRESELECT=HOTHORN or PRESELECT=LOH in PROC HPFOREST.  We did several simulations and concluded that PRESELECT=LOH is preferable. 

 

-Padraic

Contributor
Posts: 47

Re: decision stump and conditional decision trees

Posted in reply to PadraicGNeville
Hi, Padraic,
Thanks for your reply. I will follow the steps you have mentioned to see if these work for me.

Regards
Contributor
Posts: 47

Re: decision stump and conditional decision trees

Posted in reply to PadraicGNeville

Hi Padraic,

I was able to apply Decision Stump but could not find PRESELECT option in any of the node properties to get conditional decision trees.

Would you please help me about it?

 

Regards

 

SAS Employee
Posts: 37

Re: decision stump and conditional decision trees

Posted in reply to geniusgenie

Hi, GeniusGenie,

 

I also see no property in the Graphical User Interface.  The 'conditional inference' method happens automatically because it is the default in the underlying PROC HPFOREST.

 

Before EM 14.2, PRESELECT=HOTHORN was the only choice.

 

Beginning in 14.2, if there are no nominal inputs, or if all the inputs have the about the same the number of categories, then the conditional inference method is not motivated and not done.  Otherwise condition inference is done using the method of Wie-Yin Loh of University of Wisconson.  (Technical detail: 'about the same number' means within 5 categories, with non-nominal inputs treated as having one category.)

 

Regards,

Padraic

SAS Super FREQ
Posts: 306

Re: decision stump and conditional decision trees

Posted in reply to PadraicGNeville

You can set this option by running the following in the Enterprise Miner Project Start Code before running the HP Forest node:

 

%let EM_HPFOREST_PROCSTMNT = %str(PRESELECT=LOH);

Contributor
Posts: 47

Re: decision stump and conditional decision trees

Posted in reply to WendyCzika
Hi Wendy, Thanks for your message, i am able to enter this code. Could you also tell me is this option going to work for all HP Forest nodes(In case I have multiple nodes) or I can use it for a separate node as well. I have a standard HP Forest node and I am looking to add a another node separately on which I am looking to implement this code as a conditional decision tree.
So, what should I do in this case?
Ask a Question
Discussion stats
  • 6 replies
  • 447 views
  • 3 likes
  • 3 in conversation