Decision Trees and Random Forests: A Visual Introduction For Beginners: A Simple Guide to Machine Learning with Decision Trees

by Chris Smith

Step 1: Determine Your Initial Question/Problem Step 2: Determine Your Decision and Unknown Step 3: Determine the Values Step 4: Determine the Probabilities Step 5: Calculate the Weighted Value Step 6: Calculate the Net Benefit of Each Decision To

To successfully predict an outcome, a decision tree algorithm must build a tree from a set of training data, which means it must decide how to split the data.

We have only touched the surface of "purity" though, so let's spend a few moments breaking down exactly what purity is.