In Enterprise Miner, can I merge two datasets by a variable in the role of ID?

[Lychee]

I have two datasets with a commnon variable named Customer_ID. This ID variable's role was set to ID when it was initially imported into the EM as data source. Now I was trying to combine two two datasets through "Merge" node. In the property window, I set Merging type as "Match". But when I was trying to select the common varaible in the "By Ordering" menu, the Customer_ID was not there.

[WendyCzika · Accepted solution]

Before opening the By Ordering dialog, right-click on the Merge node and select Edit Variables.  Then you can set the Merge Role for Customer_ID to be "By".  After clicking OK to close this dialog, you can now open the By Ordering dialog and you should see Customer_ID show up now (this step isn't necessary though).

Hope that helps!

[Lychee]

Thank you so much, Wendy. You saved me lots of time. I spent whole day yesterday but can't figure it out.

A further question - After running the Merge node, I right-clicked on the Merge node and select Edit Variables. All the variables listed there are coming from only one of two datasets. I thought I should see the variables from both datasets.

[WendyCzika]

To see the variables in the merged data set, you can click on the button for Exported Data from the Properties panel when the Merge node is selected, then either Browse or click Properties to see the variables in the merged, exported train data.  Or connect the Merge node to another node, and in the new node you can do Edit Variables to see all the variables.