K-means++

K-means++ is a way of finding clusters in data. Think of a cluster like a group of items that are similar to each other. K-means++ helps you find the group by picking the points that are most different from the other points. For example, if you had a bunch of points on a graph, K-means++ would find the two most different points and group them together. It would then look at the remaining points and find the two that are most different and create another group. It would keep doing this until all the points are in a group.