10-02-2014 01:33 PM
Please provide a little more background.
I frequently use decision trees to model each segment of a cluster node, but the target variable is not the segment variable.
For example for the German Credit data set I can add a cluster node that will give me X segments. For each segment I add a decision tree that will model the probability of default (variable BAD). Variable bad is the target, not the variable segment created by the cluster node.
Does that help? What are you trying to do?
10-03-2014 10:23 AM
I am trying to cluster the bank customer and then I want to use tree to model their assets
What are the steps to do this, after you get the result from Cluster Node?:
For each segment I add a decision tree that will model the probability of default (variable BAD).
10-03-2014 03:29 AM
Maybe first use a Metadata node to change the ROLE of the _SEGMENT_ variable.
If I remember correctly the Cluster Node itself creates a small decision tree, where the target is the segment. You can use that tree to inspect your clusters or get approximate rules for you clusters.
10-03-2014 10:19 AM
Cluster Node itself creates a small decision tree, where the target is the segment?
I see a tree will all segment only, do you know where to get the tree for each segment individually?
10-03-2014 10:46 AM
I think there is only this tree with the _SEGMENT_ variable in the target role. No individual trees for each segment.
Sorry, previously I thought you want to create a decision tree, to explore your clusters.
If you want one model for each segment: In this case you can use the Star/End Group nodes and get a segmented model. And then put a tree node between them.
Or you want to use the _SEGMENT_ variabel as an INPUT variable? This is also possible, you need to change the ROLE of the _SEGMENT_ variable to INPUT.