Create Decision Tree in Miner with Interval Explanatory Variable

[_MVB_]

Hi everyone - hope you can help with a quick solution.

I looked through existing topics, and found relevant, but when trying to apply it in Miner, still have an issue to produce a decision tree.

here is what i try to do:

I have a dataset with binary target variable (default indicator: 1 or 0) and an interval (numeric) variable 'time on books' that goes from 0, 1, 2 ... up to ~300 and represents months on books.

In Miner, I want to build a simple decision tree that have a maximum depth: 1, and maximum branch: 3 - so just splitting 'time on books' into 3 nodes.

the issue is:

Once i set up constraints, such as branch, depth, node leaf size, etc..., and once "Run" decision tree i do not have any split for the "Input" variable, but only keeping a "root node" with information on how many are in 0 and 1 groups.

I tried "Interactive Binning" and "Transform Variables" to have this input variable either binned or transformed, but not luck.

If anyone had same issue, i must have doing something wrong in the setup, because i can create simple decision trees for other variable i have, but struggling with "time on books" and another character nominal variable.

Any help highly appreciated!

[ChrisNZ]

Moved to data mining community