Classical models built from the math you already know
Think of building a family tree for data points: every point starts alone, and the two closest relatives merge first. Start with a flat stack of name cards, no branches drawn yet, and scan every pair of cards to merge whichever two relatives are closest into one small branch. Then step back and look again: some cards are still lone individuals and some are already-joined pairs, so compare all of those and merge whichever two are closest this time, a person, a pair, or a small cluster of cousins. Keep repeating that single move, always merge the closest two, until every card has folded into one giant tree with a single root. That whole process, drawn out branch by branch, is hierarchical clustering, and the tree it produces is called a dendrogram.
🔒 This is a Pro lesson — the interactive figure, worked examples, quiz and practice open with Pro access.
▶ Hierarchical Clustering & Dendrograms