Abstract
Sometimes, we prepare a long list of grocery items and go to the supermarket well prepared to buy what we want. However, sometimes the midday sugar trigger can send us to the supermarket like a dart for picking up a chocolate. People who prepare list of items generally spend way more time in the store than those who do not. From the perspective of the store buyers, spending more time in the store is a great thing. These buyers are what we can call “Organized Buyers”. They know what they want and how much of it that they want. On the flip side, we have those buyers who just drop in the store for picking one or two items on a real physical/mental need trigger. These people are what we can call “Disorganized Buyers”. Clustering is an unsupervised machine learning technique to automatically categorize datasets like these customers/buyers are for the store. In more general terms, clustering can be thought of as automatic grouping of things, behaviors, and so on. There is obviously a known right answer to the number of groups present in a dataset, but it is impossible to be known for each and every dataset in prior.