Clustering is an unsupervised learning method that groups similar observations into clusters, without any prior labels. It is used to segment, explore, or summarize data. The result depends on the distance measure, the scale of the variables, and the chosen algorithm.