According to Han (2011), Clustering is the process of grouping a set of data objects into multiple groups or clusters so that objects within a cluster have high similarity, but are very dissimilar to objects in other clusters (p. 443).
Imagine a dinner party has just started. You see a medium-sized rectangular room with a number of people in it. The room is filled with people who are socializing. You notice that people have formed small groups. What draws them together? It could be their existing friendship. Perhaps it is a future networking opportunity. Some groups may form for less obvious reasons. One thing you can say is that it would be unlikely to see everyone in the center of the room talking together as a single group of people. Keep this mental image in your mind because it represents an underlying aspect of cluster analysis.
Clusters are collections of points from a multidimensional set of data such that they minimize the distance between each cluster...