Chapter 13. K-means

In previous chapters, we discussed supervised learning tasks; we examined algorithms for regression and classification that learned from labeled training data. In this chapter, we will introduce our first unsupervised learning task: clustering. Clustering is used to find groups of similar observations within a set of unlabeled data. We will discuss the K-means clustering algorithm, apply it to an image compression problem, and learn to measure its performance. Finally, we will work through a semi-supervised learning problem that combines clustering with classification.