Introducing Machine Learning

Chapter 15. How to Group Data: Classification and Clustering

If a machine is expected to be infallible, it cannot also be intelligent.

—Alan Turing, from “Computing Machinery and Intelligence”, 1950

In the previous chapters, you encountered the term classification many times. Classification is one of the most common problems in machine learning, and it can be tackled in various ways, including decision trees, Bayesian classifiers, and even logistic regression.

In this chapter, we’ll present two more sophisticated algorithms for classification, and then we’ll move on to address a subtly similar problem—clustering. According to most dictionaries, classification is the act of arranging a group of things in homogeneous classes based on their characteristics. ...