From the course: Machine Learning and AI Foundations: Decision Trees with SPSS

Unlock the full course today

Join today to access over 22,700 courses taught by industry experts or purchase this course individually.

A quick look at the complete C&RT tree

A quick look at the complete C&RT tree

From the course: Machine Learning and AI Foundations: Decision Trees with SPSS

Start my 1-month free trial

A quick look at the complete C&RT tree

“

- [Instructor] We're looking at the entire C&RT tree. This is so dramatically different than the CHAID tree that we probably begin by stating the obvious, it's a much smaller tree. I want you to recall something, though, about our discussion of the analysis node. Recall that initially C&RT looked like it might be somewhat less accurate than CHAID, but it was stable, where the performance on the train data set and the test data set were similar. Whereas in CHAID, there was a dramatic drop in performance between the train data set and the test data set and that made us nervous. It made us concerned that perhaps that tree was overfit. Well you're looking at a tree that would exhibit those behaviors in an analysis node. There's relatively few leaf nodes at all, and there's only one leaf node with a small sample size. Node five has a sample size of 12, so we would expect a tree like this to be very different. If anything, in fact, we might be tempted to change the settings and experiment…

Contents