FactoClass: Combination of Factorial Methods and Cluster Analysis

Some functions of 'ade4' and 'stats' are combined in order to obtain a partition of the rows of a data table, with columns representing variables of scales: quantitative, qualitative or frequency. First, a principal axes method is performed and then, a combination of Ward agglomerative hierarchical classification and K-means is performed, using some of the first coordinates obtained from the previous principal axes method. The function 'kmeansW', a modification of 'kmeans', programmed in C++, is included, in order to permit to have different weights of the elements to be clustered. Some complementary functions and datasets are included. See, for example: Lebart, L. and Piron, M. and Morineau, A. (2006). Statistique Exploratoire Multidimensionnelle, Dunod, Paris.

Downloads:

Reverse dependencies:

Linking:

Please use the canonical form https://CRAN.R-project.org/package=FactoClass to link to this page.