perturb.by.noise: Function to generate a data set perturbed by noise.

In mosclust: Model Order Selection for Clustering

Function to generate a data set perturbed by noise.

Description

This funtion adds gaussian noise to the data. The mean of the gaussian noise is 0 and the standard deviation is estimated from the data. The gaussian noise added to the data has 0 mean and the standard deviation is estimated from the data (it is set to a given percentile value of the standard deviations computed for each variable).

Usage

perturb.by.noise(X, perc = 0.5)

Arguments

X	matrix of data (variables are rows, examples columns)
perc	percentile of the standard deviation (def: 0.5)

Value

matrix of perturbed data (variables are rows, examples columns)

Author(s)

Giorgio Valentini [e-mail]

References

McShane, L.M., Radmacher, D., Freidlin, B., Yu, R., Li, M.C. and Simon, R., Method for assessing reproducibility of clustering patterns observed in analyses of microarray data, Bioinformatics, 11(8), pp. 1462-1469, 2002.

See Also

do.similarity.noise

Examples

library("clusterv")
# Data set generation
M <- generate.sample6 (n=20, m=10, dim=100, d=3, s=0.2);
# generation of a data set perturbed by noise
M.perturbed <- perturb.by.noise(M);
# generation of a data set more perturbed by noise
M.more.perturbed <- perturb.by.noise(M, perc=0.95);

mosclust documentation built on June 8, 2025, 11:23 a.m.