A decision tree makes predictions by asking a series of simple questions about the data, splitting it into smaller and smaller groups. Trees are easy to read and explain, which makes them one of the most popular building blocks in machine learning.
You'll start with classification trees and how a tree chooses its splits, then see how trees handle feature selection and missing data. Next come regression trees for predicting numbers, and finally pruning, which keeps a tree from overfitting by cutting back branches that don't help on new data.
Watch the videos in order, then try building and pruning a tree on a dataset of your own. When you're done, continue on to Ensembles: Forests & Boosting.
How a classification tree is built, and how it uses impurity to decide which question to ask at each split.
How trees pick the most useful features, and how they cope with missing values and numeric data.
How trees predict continuous values by splitting data to minimize squared error.
How cost complexity pruning trims a tree to avoid overfitting, with cross-validation to choose the penalty.