2024-03-29T10:17:44Zhttps://repositori.uji.es/oai/requestoai:repositori.uji.es:10234/1163822023-02-15T20:05:00Zcom_10234_7037com_10234_9col_10234_8635
00925njm 22002777a 4500
dc
Vinué Visús, Guillermo
author
Epifanio, Irene
author
Alemany Mut, Sandra
author
2015
The new concept archetypoids is introduced. Archetypoid analysis represents each observation in a dataset as a mixture of actual observations in the dataset, which are pure type or archetypoids. Unlike archetype analysis, archetypoids are real observations, not a mixture of observations. This is relevant when existing archetypal observations are needed, rather than fictitious ones. An algorithm is proposed to find them and some of their theoretical properties are introduced. It is also shown how they can be obtained when only dissimilarities between observations are known (features are unavailable). Archetypoid analysis is illustrated in two design problems and several examples, comparing them with the archetypes, the nearest observations to them and other unsupervised methods.
0167-9473
http://hdl.handle.net/10234/116382
http://dx.doi.org/10.1016/j.csda.2015.01.018
Archetype
Convex hull
Unsupervised learning
Extremal point
Non-negative matrix factorization
Archetypoids: A new approach to define representative archetypal data