cutCalculation: cutCalculation function

In uHMM: Construct an Unsupervised Hidden Markov Model

Description

Compute intra and inter-cluster cuts from the similarity matrix of a dataset.

Usage

cutCalculation(similarity, label, K)

Arguments

similarity	a similarity matrix.
label	vector of cluster sequencing.
K	number of clusters. (= nbCluster CALCULE DANS LA FONCTION ???)

Details

intra cluster cut :

Cut(g_{k},g_{l}) = ∑_{i=1,x(i)\in g_{k}}^{N_{p}}∑_{j=1,x(j)\in g_{l}}^{N_{p}}w(x(i),x(j))

Value

The function returns a list containing:

mncut	the inter-cluster cut, i.e. K-sum(ratioCutVol).
ratioCutVol	vector of intra-cluster cuts, one component per cluster.

Examples

x<-rbind(matrix(runif(100),ncol=2),matrix(runif(100)+2,ncol=2),matrix(runif(20)*3,ncol=2))
similarity<-ZPGaussianSimilarity(x,7)%*%t(ZPGaussianSimilarity(x,7))
km<-kmeans(similarity,2)
label<-km$cluster
plot(x,col=km$cluster)
cutCalculation(similarity,label,length(unique(label)))

uHMM documentation built on May 2, 2019, 9:16 a.m.