Archetypal analysis in

statistics Statistics (from German language, German: ''wikt:Statistik#German, Statistik'', "description of a State (polity), state, a country") is the discipline that concerns the collection, organization, analysis, interpretation, and presentation of ...

is an unsupervised learning method similar to

cluster analysis Cluster analysis or clustering is the task of grouping a set of objects in such a way that objects in the same group (called a cluster) are more similar (in some sense) to each other than to those in other groups (clusters). It is a main task of ...

and introduced by Adele Cutler and

Leo Breiman Leo Breiman (January 27, 1928 – July 5, 2005) was a distinguished statistician at the University of California, Berkeley. He was the recipient of numerous honors and awards, and was a member of the United States National Academy of Sciences. ...

in 1994. Rather than "typical" observations (cluster centers), it seeks extremal points in the multidimensional data, the " archetypes". The archetypes are convex combinations of observations chosen so that observations can be approximated by convex combinations of the archetypes.

Literature

* Adele Cutler and Leo Breiman. Archetypal analysis. Technometrics, 36(4):338–347, November 1994.
Manuel J. A. Eugster: Archetypal Analysis, Mining the Extreme. HIIT seminar, Helsinki Institute for Information Technology, 2012
* Anil Damle, Yuekai Sun: A geometric approach to archetypal analysis and non-negative matrix factorization. arXiv preprint
arXiv : 1405.4275
Data mining Cluster analysis