In this section we introduce the problem of clustering, that is, how to create meaningful groups from unlabeled datasets. When creating different clusters, we will aim to have similar data in the same group (cluster) and different ones in different groups, so we analyze how the concept of similarity and dissimilarity can be included mathematically in the clustering methodologies.

In this first subsection, we introduce the concept of clustering and similarity and dissimilarity. After studying this subsection, you should be able to:

• Explain the problem of clustering.
• Understand different measures of similarity and dissimilarity in clustering.