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

use Confusion matrix as the evaluation criterion in Enterprise Miner

Reply
Learner
Posts: 1

use Confusion matrix as the evaluation criterion in Enterprise Miner

I got a question about Enterprise Miner. Is it possible to use confusion matrix as the main evaluation criterion to select the champion model when doing it with the model comparision node? I'd like to use cost 5 for false negative and cost 1 for false positive.

 

Thank you

SAS Employee
Posts: 340

Re: use Confusion matrix as the evaluation criterion in Enterprise Miner

Have you tried to set decision weights (profit matrix)? You can do it while importing the data or there is a dedicated node for it.
If you do this, Model Comparision Node can use those weights.
Super Contributor
Posts: 336

Re: use Confusion matrix as the evaluation criterion in Enterprise Miner

As usual, Gergely has good advice! Another way to assign cost or profits is to use the Decisions node in the assessment tab.

I hope this helps!

Ask a Question
Discussion stats
  • 2 replies
  • 671 views
  • 1 like
  • 3 in conversation