Normlized data in classification trees

I am applying a decision tree learner. Before that, I apply Normalizer to normalize the data, and partition it in two partitions. Finally, I am applying the learned tree on the predictor, with the second partition as entry.

When I am analyzing my decision tree, the values of the decision points are also normalized. This is not useful, because I would like to see the non-normalized values displayed to interpret the tree. Is there a way in Knime to achieve this, i.e. on the basis of learned normalized decisions tree to have a decision tree that shows the non-normalized values?

Thanks for the help

Is this post helpful;


If I understand well, the solutions proposed on the referenced post apply on the resulting table. Isn't it possible to apply a denormalization on the model itself, rather than on the result of the model?

I guess you don't need to normalize your data, at least not when used in the decision tree. The output is equivalent. (Decistion tree algorithms look at each attribute independently -- and then it doesn't make a difference whether the data is normalized or not.)