How to bin optimally to receive optimal weight of evidence (WOE) in KNIME?

I’m trying to build a scorecard and I need to find the optimal binning of the scale variables. The CAIM doesn’t seem to perform that so far.
Any suggestions?

Hi @deicide_bg -

Sorry for the delayed reply here. I know that Dymatrix has a commercial KNIME extension that supports scorecard building, and in particular, discrete binning using WOE. See their presentation from the 2014 KNIME User’s Group Meeting. That said, I don’t know if those nodes are in active development any more.

The other way to go about this would be with R or Python, I’m afraid.

3 Likes

This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.