Clustering Lab · Playground · Lorenzo Marinelli

Algorithm

Settings

Clusters 3 Iterations 16

Points

Ready

Algorithm k-means

Points 0

Clusters 0

Noise 0

k-means forces every point into one of k compact groups.

Repeatedly assigns each point to the nearest center, then moves each center to the mean of its assigned points.

Best when: Groups are compact and roughly round.

Watch out: The cluster count is forced even when the geometry disagrees.