Hierarchical Clustering

Harvard EPS-210 | Interactive tutorial — Explore agglomerative clustering and dendrograms

Clustering Space

Unclustered
Merge Step

Distance Matrix

Dendrogram

Drag the cut height slider to select number of clusters

Data Points

Click on the clustering canvas to add points

Cut Height

--
Clusters
Cut Height 0.00

Linkage Method

d(A,B) = avg d(a,b)
Average of all pairwise distances

Algorithm Control

Statistics

Points
0
Merges
0
Merge history will appear here

Quick Examples

How it works: Agglomerative clustering starts with each point as its own cluster, then repeatedly merges the two closest clusters until one remains. The dendrogram shows this hierarchy.