Affinity Propagation

Affinity propagation performs the clustering by identifying exemplars (also called prototypes) among the available data points and reports their neighborhoods as clusters. It considers all data points as possible exemplars. The final set of exemplars is determined by sending messages between the data points voting for the best set of exemplars for the given similarity function in combination with the preference.

• Publication: Brendan J Frey and Delbert Dueck. Clustering by passing messages between data points. Science, 315(5814):972–976, Feb 2007.