Affinity propagation — how does it work?

hard

Answer

$Message-passing algorithm exchanging 'responsibility' and 'availability' messages between all pairs of points until exemplars emerge.$
$No k needed — 'preference' parameter influences the number of clusters.$
$O(n^{2})$
$Use for small-to-medium exploratory clustering when k is unknown and interpretable exemplars matter (e.g., customer archetypes).$

Check yourself — multiple choice

Random
Message-passing exchanges responsibility + availability; auto-selects exemplars; no k needed but $O(n^{2})$ ; interpretable exemplars
Same as k-means
Not real

Affinity propagation: message-passing → exemplars; no k needed; $O(n^{2})$ .

#clustering

214 interview questions in this topic.