The website is in Maintenance mode. We are in the process of adding more features.
Any new bookmarks, comments, or user profiles made during this time will not be saved.

AIML.com

Machine Learning Resources

How does K-Means ++ work?

Bookmark this question

K-Means ++ has been generally shown to be the best initialization approach to use when performing K-Means clustering. At a high level, it seeks to maximize the distance between points chosen as the initial cluster centroids. As one of the goals of clustering is to find clusters that distinguish between observations, it makes sense to start with centroids that are as far apart as possible. It does this by randomly choosing one data point to be the first centroid and then computing a probability for each of the other observations to be chosen in a way that the maximum probability is given to observations furthest apart from the centroids already chosen until k such observations are found. The K-Means algorithm proceeds as usual once the initial centroids have been chosen. 

Leave your Comments and Suggestions below:

Please Login or Sign Up to leave a comment

Partner Ad  

Find out all the ways
that you can

Explore Questions by Topics