yaConsensus (version 1.1)

consensus.diss: Computes the consensus dissimilarity matrix.

Description

Computes the consensus dissimilarity according to the algorithm of Monti et al. (2003).

Usage

consensus.diss(cclusters, similarity = FALSE)

Value

An object of the 'dist' class.

Arguments

cclusters

a matrix of integers where the column are the samples, and the rows are different clusterings of the samples.

similarity

a logical value signaling if the similarity matrix is required.

Author

Stefano M. Pagnotta

Details

In any row of the ccluster matrix, the value 0 means that the corresponding sample is not assigned to any cluster. In this case, the dissimilarity is computed consistently.

References

Monti et al. (2003) - Consensus Clustering: A Resampling-Based Method for Class Discovery and Visualization of Gene Expression Microarray Data - Machine Learning 52(1-2):91-118 <DOI: 10.1023/A:1023949509487>

See Also

dist

Examples

clusters <- rep(1:3, c(3, 9, 18))
clusterings <- matrix(NA, ncol = 30, nrow = 50)
for(k in 1:50) clusterings[k,] <- sample(clusters)
ddist <- consensus.diss(clusterings)
class(ddist)
attr(ddist, "method")