diam_clus: Diametrical clustering

In watson: Fitting and Simulating Mixtures of Watson Distributions

Diametrical clustering

Description

diam_clus clusters axial data on sphere using the algorithm proposed in Dhillon et al. (2003).

Usage

diam_clus(x, k, niter = 100)

Arguments

x	a numeric data matrix, with rows corresponding to observations. Can be a dense matrix, or any of the supported sparse matrices from Matrix package by Rcpp.
k	an integer giving the number of mixture components.
niter	integer indicating the number of iterations of the diametrical clustering algorithm, default: 100.

Value

a matrix with the concentration directions with an attribute "id" defining the classified categories.

References

Inderjit S Dhillon, Edward M Marcotte, and Usman Roshan. Diametrical clustering for identifying anti-correlated gene clusters. Bioinformatics, 19(13):1612-1619, 2003.

Examples

## Generate a sample 
a <- rmwat(n = 200, weights = c(0.5,0.5), kappa = c(20, 20),
                        mu = matrix(c(1,1,-1,1),nrow = 2))
## Fit basic model
q <- diam_clus(a, 2)

watson documentation built on April 12, 2025, 1:35 a.m.