Tree-Based Machine Learning Methods in SAS Viya

Chapter 3: Growing a Decision Tree

Recursive Partitioning

Recursive partitioning repeatedly splits parent nodes into child nodes to create a decision tree. It’s a simple three-step procedure shown in Figure 3.1.

An easy algorithm first splits the entire data set, the root node, into at least two nodes. Each node is then split into more nodes, and so on. Each node in its turn is considered in isolation. The search for a split is done without any concern as to how another node is to be split. This process is called recursive partitioning and the majority of tree algorithms use it.

The algorithm splits the parent node into two or more child nodes in such a way that the values (or levels) of the response variable within each child region are as similar ...