k-Medoids (PAM)

In contrast to , Partitioning Around Medoids (PAM) is one of the most common realizations of the k-medoid clustering approach, which uses exemplars as cluster centers. The algorithm works identical to k-means except that the exemplars are updated by testing each object in each cluster as exemplar instead of calculating artificial mean objects (centroids).

• Publication: Martin Maechler, Peter Rousseeuw, Anja Struyf, Mia Hubert, and Kurt Hornik. cluster: Cluster Analysis Basics and Extensions, 2012. R package version 1.14.3 — For new features, see the ’Changelog’ file (in the package source).