Smart Data Discovery Using SAS Viya

Chapter 8: Predictive Modeling Using Decision Trees

Overview

Model Building and Assessment

Data Preparation

Predictor Variables Selection

Building the Decision Tree Model

Tree Window

Interpreting the Model

Overall Fit and Assessment Plots

Misclassification Plot

Variable Importance Chart

Lift Chart

Tuning and Improving the Model

Adding and Removing Predictor Variables

Dealing with Missing Values

Tuning the Hyperparameters

Tree Pruning

Overview

When it comes to building predictive models, it is often best to try the simplest approach first. More sophisticated techniques and algorithms are attractive, but they are usually harder to use since they typically require more configuration and fine-tuning in order to get good results. When it comes to ...