EK: Estimate the size of cluster for continuous data.

In saeidamiri1/GHC: General Hybrid Clustering Technique

Usage

EK(x,B=200,knmin,knmax)

Arguments

x	An nxp data matrix; n is the number of observations and p is the number of dimensions.

B	Number of iterations.
knmin	Minimum size of cluster to get the stabilized clustering, default knmin=2
knmax	Maximum size of cluster to get the stabilized clustering, default knmax=n/5

Details

The function 'EK' gestimate the size of cluster presented in Algorithm 2, see Amiri et al. (2018) for details.

Author(s)

Saeid Amiri, Bertrand Clarke and Jennifer Clarke.

References

Amiri, S., Clarke, B, Clarke, J. & Koepke, H.A. (2018). A General Hybrid Clustering Technique. To appear in Journal of Computational and Graphical Statistics.

Examples

library("foreach")
library("doParallel")

knmin0<-2
knmax0<-floor(dim(spiral)[1]/5)

KCLUS<-EK(spiral,B=200,knmin=knmin0,knmax=knmax0)

# plot the dendrogram
plot(hclust(KCLUS[[1]],method="single"),h=-1)

# print the assigned clusters to observation
print(KCLUS[[2]])

saeidamiri1/GHC documentation built on May 22, 2019, 2:20 p.m.