In this section, you'll learn about a useful unsupervised learning technique: clustering. This lesson summarizes the topics you'll be covering in this section.

Clustering techniques are very powerful when you want to group data with similar characteristics together, but have no pre-specified labels. The main goal of clustering is to create clusters that have a high similarity between the data belonging to one cluster while aiming for minimal similarity between clusters.

We start by providing a basic intuition of the K-means clustering algorithm. When using the K-means clustering algorithm, the number of clusters that you want to obtain is specified upfront and the algorithm aims at the most "optimal" cluster centers, given that there are K clusters.

A second branch of clustering algorithms is hierarchical agglomerative clustering. Using hierarchical clustering, unlike K-means clustering, you don't decide on the number of clusters beforehand. Instead, you start with n clusters, where n is the number of data points, and at each step you join two clusters. You stop joining clusters when a certain criterion is reached.

Semi-supervised learning techniques, which are increasingly popular in machine learning, combine both concepts of supervised and unsupervised learning.

A very common and useful application of clustering is market segmentation. You'll practice your clustering skills on a market segmentation dataset!

In this section, you'll learn how to use clustering techniques which are very useful for finding patterns and grouping unlabeled data together.