kmed (version 0.1.0)

rankkmed: Rank k-medoid algorithm from Zadegan et. al.

Description

This function computes and returns the clustering result computed by using a specified distance via rank k-medoids algorithm.

Usage

rankkmed(distdata, ncluster, m = 3, iterate = 10, initial = NULL)

Arguments

distdata

A matrix of distance objects (n x n) or a diss class.

ncluster

A number of cluster.

m

A number of objects to compute hostility.

iterate

A number of iteration for clustering algorithm.

initial

A vector of initial objects as the cluster medoids.

Value

Function returns a partitioning clustering algorithm result consists of cluster membership, cluster medoid, the minimum distance to the cluster medoid.

Details

This is a k-medoids algorithm that has been proposed by Zadegan et. al. The algorihm has been claimed to be suitable for large dataset. The medoids updating in this algorihm is similar to kmeans centroid updating.

References

Zadegan, S.M.R, Mirzaie M, and Sadoughi, F. 2013. Ranked k-medoids: A fast and accurate rank-based partitioning algorithm for clustering large datasets. Knowledge-Based Systems 39, 133-143.

Examples

# NOT RUN {
num <- as.matrix(iris[,1:4])
mrwdist <- distNumeric(num, num, method = "mrw")
result <- fastkmed(mrwdist, ncluster = 3, iterate = 50)
table(result$cluster, iris[,5])


# }