rapid miner: how to add a 'label' attribute to a dataset?

fstab picture fstab · Apr 8, 2013 · Viewed 25.6k times · Source

I want to apply a decision tree learning algorithm to a dataset I have imported from a CSV. The problem is that the "tra" input of the Decision Tree block is still red, stating "Input example set must have special attribute 'label'.". How do I add that label? What is it? I have been playing around with it for some time without results. Any help appreciated.

ADDENDUM: the column's titles have been correctly inferred, so I have no clue on what the 'label' is.

Answer

Mayank Agarwal picture Mayank Agarwal · Jan 28, 2014

When you import data in RapidMiner, In Step number 4, you need to select the attribute type "label" for the column you wish to classify.

Labelling the classifier column as with "label" attribute

Click Next button in the bottom.

In Step 5, it will ask you to save the import file, Choose any suitable name and click 'Finish'.