In this chapter, we get into machine learning and how to actually implement machine learning models in Python.

We'll examine what supervised and unsupervised learning means, and how they're different from each other. We'll see techniques to prevent overfitting, and then look at an interesting example where we implement a spam classifier. We'll analyze what K-Means clustering is a long the way, with a working example that clusters people based on their income and age using scikit-learn!

We'll also cover a really interesting application of machine learning called decision trees and we'll build a working example in Python that predict shiring decisions in a company. Finally, we'll walk through the fascinating concepts of ensemble learning and SVMs, which are some of my favourite machine learning areas!

More specifically, we'll cover the following topics:

• Supervised and unsupervised learning
• Avoiding overfitting by using train/test
• Bayesian methods
• Implementation of an e-mail spam classifier with Naïve Bayes
• Concept of K-means clustering
• Example of clustering in Python
• Entropy and how to measure it
• Concept of decision trees and its example in Python
• What is ensemble learning
• Support Vector Machine (SVM) and its example using scikit-learn