Let's get started. In this section, we will contrast unsupervised learning with supervised learning—the latter is what we have been doing up to this point. We will then learn what clustering is.
Supervised and unsupervised learning are two different learning paradigms. So far in this book, we have been performing different kinds of supervised learning. There was a variable, which we called the target variable, that we wanted to predict. This was the case both for classification and regression. Learning consisted of finding a classifier that could accurately predict the target variable. The following diagram represents supervised learning:
With unsupervised learning, there is no target variable. We believe there are hidden features that differentiate the data, but we don't observe them. The objective of unsupervised learning is to find...
