mlcc.kmeans: Multiple Latent Components Clustering - kmeans algorithm

In psobczyk/public_varclust: Variable Clustering

Description

Performs k-means based subspace clustering. Center of each cluster is some number of principal components. Similarity measure is R^2 coefficient.

Usage

mlcc.kmeans(X, number.clusters = 2, stop.criterion = 1, max.iter = 40,
  max.subspace.dim = 4, initial.segmentation = NULL,
  estimate.dimensions = FALSE)

Arguments

X	a matrix with only continuous variables
number.clusters	an integer, number of clusters to be used
stop.criterion	an integer indicating how many changes in partitions triggers stopping the algorithm
max.iter	an integer, maximum number of iterations of k-means
max.subspace.dim	an integer, maximum dimension of subspaces
initial.segmentation	a vector, initial segmentation of variables to clusters
estimate.dimensions	a boolean, if TRUE (value set by default) subspaces dimensions are estimated

Value

A list consisting of:

segmentation	a vector containing the partition of the variables
pcas	a list of matrices, basis vectors for each cluster (subspace)

psobczyk/public_varclust documentation built on May 26, 2019, 10:33 a.m.