What is hierarchical clustering and when is it useful?
mediumAnswer
- It builds a tree (dendrogram) of nested clusters, either agglomerative (bottom-up: merge closest clusters) or divisive (top-down).
- You cut the tree at a chosen height to get k clusters.
- Useful when you want to explore cluster structure at multiple scales, or when you don't know k in advance and want a visual.
- Downside: memory, so it doesn't scale beyond ~10k points.
Check yourself — multiple choice
- It scales linearly to millions of points
- It produces a dendrogram and works well on smaller datasets
- It requires k in advance
- It only supports divisive splitting
Hierarchical clustering yields a dendrogram; memory limits scale.
#clustering
Practise Unsupervised Learning
214 interview questions in this topic.