Derive – Nominal

One limitation of deriving a field as a flag is that you can only create a field with two outcomes. Deriving a field as a Nominal allows you to create a field with many categories. One of the most common uses of deriving a field as a Nominal is to transform a continuous field into a categorical field with multiple groups (for example, age groups, income groups, and so on).

Note

It is common in data mining to create multiple versions of a field to see if one version is a better predictor in a model. If we wanted to create categories that were of equal width, equal size, or were based on standard deviations, the Binning node could be used instead.

In this example, we will modify the Age variable by classifying Age into a categorical variable called Age_Groups, which is the Age field banded into six groups, Young, Thirties, Forties, Fifties, Sixties, and Retired:

1. Place a Derive node onto the canvas.
2. Connect the Derive node named Employed to the new Derive node.
3. Edit the new Derive...