Tree-Based Machine Learning Methods in SAS Viya

Chapter 2: Classification and Regression Trees

Classification Trees

A classification tree is a predictive model where the target variable is categorical. The decision tree algorithm is then used to identify the “class” or “label” within which a target variable would most likely fall. An example of a classification-type problem would be determining who will or will not default to a bank loan or who will or will not churn from a telecom service provider.

Consider the case of a nominal target variable. Handwriting recognition is a classic application of supervised prediction.

The example data set is a subset of the pen-based recognition of handwritten digits data, available from the UCI repository (Blake et al. 1998). The cases are digits written ...