Grouping related samples together is generally used in unsupervised learning. Once all samples are grouped, the researcher can optionally assign meaning to each cluster.

There are many clustering algorithms. For example, the k-means algorithm clusters samples based on the proximity of the samples to the centroid, as shown in the following figure:

Researchers can then view these clusters and perform other operations, such as labeling cluster 1 as "dwarf trees" and cluster 2 as "full-sized trees."