Clustering by k-means

Clustering is an example of unsupervised learning, where a set of data are partitioned into a certain number of groups. K-means is one of the simplest methods of clustering.

Below, you can interactively input 2D data points, and see the result of k-means clustering.

  • Clicking on the area adds a data point.
  • Clicking while hoding down an alt key removes the point under cursor.
  • Hold down a shift key to start dragging the point under cursor.

'X's indices the means of the clusters.