Kmedoids: Perform k-medoids clustering on a data matrix. After...

In neurodata/knorR: A Parallel Clustering Non-Uniform Memory Access ('NUMA') Optimized Library

Description

Perform k-medoids clustering on a data matrix. After initialization the k-medoids algorithm partitions data by testing which data member of a cluster Ci may make a better candidate as medoid (centroid) by reducing the sum of distance (usually taxi), then running a reclustering step with updated medoids.

Usage

Kmedoids(data, centers, nrow = -1, ncol = -1,
  iter.max = .Machine$integer.max, nthread = -1, init = c("forgy"),
  tolerance = 1e-06, dist.type = c("taxi", "eucl", "cos"))

Arguments

data	Data file name on disk or In memory data matrix
centers	The number of centers (i.e., k)
nrow	The number of samples in the dataset
ncol	The number of features in the dataset
iter.max	The maximum number of iteration of k-means to perform
nthread	The number of parallel threads to run
init	The type of initialization to use c("forgy")
tolerance	The convergence tolerance
dist.type	What dissimilarity metric to use

Value

A list containing the attributes of the output of kmedoids. cluster: A vector of integers (from 1:k) indicating the cluster to which each point is allocated. centers: A matrix of cluster centres. size: The number of points in each cluster. iter: The number of (outer) iterations.

Author(s)

Disa Mhembere <[e-mail]>

Examples

iris.mat <- as.matrix(iris[,1:4])
k <- length(unique(iris[, dim(iris)[2]])) # Number of unique classes
km <- Kmedoids(iris.mat, k)

neurodata/knorR documentation built on May 25, 2019, 10:35 p.m.