QuestionQ29

VerifiedAdvanced Analytics - Theory, Application, and Interpretation of Results for Eight Methods

You create a decision tree to classify five distinct customer types from the browsing histories of a sample of 500 customers. The resulting decision tree contains 17 layers, and one leaf node contains only three customers.

What can you conclude?

Explanation

A deep decision tree that produces very small leaf nodes can fit noise and unusual patterns in the training sample instead of learning patterns that generalize to new customers. Seventeen layers and a leaf with only three customers therefore indicate a potential overfitting problem.

Community Discussion

No comments yet. Be the first to start the discussion!