Unsupervised Learning
Hierarchical Clustering
7 min+65 XPStep 1/10
Clustering without picking k
Agglomerative clustering starts with every point as its own cluster, then repeatedly merges the two closest clusters until only one remains.
The full merge history is recorded in a dendrogram — a family tree of your data. Cut the tree horizontally at any height and you get a different number of clusters: every k explored in a single run.
insight
K-means makes you guess k and pray. A dendrogram shows you the structure at every granularity first — then you decide where to cut.