Hierarchical Clustering | Interactive Workspace

Watch Hierarchical Clustering learn

Follow each update in real time. Scrub or play to inspect the algorithm.

Step 1 of 15

02 Interactive demo

CURRENT LOSS · distance

—

Merge distance at each step (dendrogram height)

WHAT'S HAPPENING

1. The Initial Dataset (15 Clusters)

Agglomerative hierarchical clustering begins with every sample as its own cluster (leaves of the dendrogram). In each step, the two closest clusters merge and add a horizontal rung at distance height d. Strategy: Single Linkage (Min Distance).

\ ext{Initial: } 15 \ ext{ singleton clusters, Dendrogram Height } d = 0

PARALLEL DENDROGRAM EXPLORER

Parallel Dendrogram Construction

Synchronized step-by-step with 2D cluster plot

Step 0 of 14

As you play or step the simulation, watch each U-branch rise up at its exact merge distance (height). Click any branch or leaf to inspect.

INITIAL STATE (15 CLUSTERS)

Every point begins in its own isolated cluster. Step forward or hit Play to observe the closest pairs merge bottom-up.