Data Science: Machine Learning - HarvardX: PH125.8x

Section 5: Classification with More than Two Classes and the Caret Package

In the Classification with More than Two Classes and the Caret Package section, you will learn how to overcome the curse of dimensionality using methods that adapt to higher dimensions and how to use the caret package to implement many different machine learning algorithms.

After completing this section, you will be able to:

• Use classification and regression trees.
• Use classification (decision) trees.
• Apply random forests to address the shortcomings of decision trees.
• Use the caret package to implement a variety of machine learning algorithms.

This section has three parts:

1. Classification with more than two classes
2. The Caret Package
3. Set of exercises on the Titanic

5.1 Classification with more than two classes

5.1.1 Trees Motivation

5.1.2 Classification and Regression Trees (CART)

5.1.3 Classification (Decision) Trees

5.1.4 Random Forests

5.1.5 Comprehension Check: Trees and Random Forests

5.2 The Caret Package

5.2.1 The Caret Package

5.2.2 Tuning Parameters with Caret

5.2.3 Comprehension Check: Caret Package

5.3 Set of exercises on the Titanic

5.3.1 Titanic Exercises, Part 1

5.3.2 Titanic Exercises, Part 2