Machine learning visualisations

Visualisations About

All visualisations/Unsupervised learning

How does k-means decide where clusters belong?

Difficulty: Approachable
Estimated time: 5 min

Loading visualisation

Step 1 of 4Place the centroids

Drag a centroid, or focus the chart and use the number keys to select one and the arrow keys to move it.

Nothing is assigned yet. The three centroids start away from the true centre of each group of points.

Drag a centroid to move it, or use number keys 1 to 3 to select a centroid and the arrow keys to nudge it. Hold Shift for larger moves.

Iteration 1, ready to assign points. Centroid 1 selected.

k