Hello,
I wanted to share my experience with you regarding output generated by the Scorecard Node in SAS Enterprise Miner (EM), Version 14.1 in the Linux environment. I noticed that EM uses labels (as opposed to variable names) when creating the output from the Scorecard Node. If you have two characteristics, say Score_t0 (for Feb 2017) and Score_t1 (for Mar 2017) for a customer account but these characteristics have the same labels, say, Score then EM may bin these two characterstics using the Score label and you may not get the correct attributes. In other words, the attributes for these two different characteristics will be grouped together as one and you may get the same group for two levels of these characteristics. I should mention that the attibutes are correct when you view the Interactive Grouping Node window but the output from the Scorecard Node is not correct.
I had the same label for this variable, since I was extracting data in Base SAS, using the SQL pass through facility from the data warehouse. For coding purposes in Base SAS, we use variable names and not labels, and so it never occured to me that EM would use labels in the Scorecard Node output. To illustrate this issue, please see the partial output that is attached.
Based on my experience, I would suggest that SAS EM should use variable names (to be consistent with Base SAS) when creating outputs in SAS EM. If other users have also experienced this issue, can we make a recommendation to the SAS Institute to use variable names and not labels for output generated in SAS EM.
Thanks,
Nilotpal
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.