Cluster analysis is a statistical classification technique in which a set of objects or points with similar characteristics are grouped together in clusters. It encompasses a number of different algorithms and methods that are all used for grouping objects of similar kinds into respective categories. The aim of cluster analysis is to organize observed data into meaningful structures in order to gain further insight from them.
Cluster analysis can be considered a tool for exploratory data analysis that is aimed at sorting different objects into meaningful groups in such a way that the degree by which these objects are associated is at the maximum if they belong to the same group and at the minimum if they do not. Cluster analysis is used to discover the hidden structures or relationships within data without having the need to explain or interpret what this relationship is. In essence, cluster analysis is only used to discover the structures found in data without explaining why those structures or relationships exist.
Cluster analysis is often applied to very simple things without us knowing it, such as food groupings at the grocery store, or a group of people eating together in a restaurant. In the grocery store, foods are grouped according to their type such as beverages, meat and produce; already, we can draw out patterns with regard to those groupings.