Artificial Intelligence By Example

The k-means clustering program

k-means clustering is a powerful unsupervised learning algorithm. We often perform k-means clustering in our lives. Take, for example, a lunch you want to organize for a team of about 50 people in an open space that can just fit those people. It will be a bit cramped but a project must be finished that day and it is a good way to bind the team for the final sprint.

A friend and a friend first decide to set up a table in the middle. Your friend points out that the people in that room will form a big cluster k, and with only one table in the geometric center (or centroid) c, it will not be practical. The people near the wall will not have access to the main table.

You have an idea. You call Pert, who runs a computation ...