Learn R Programming

⚠️There's a newer version (0.4.2) of this package.Take me there.

kmed (version 0.3.0)

Distance-Based K-Medoids

Description

Algorithms of distance-based k-medoids clustering: simple and fast k-medoids, ranked k-medoids, and increasing number of clusters in k-medoids. Calculate distances for mixed variable data such as Gower, Podani, Wishart, Huang, Harikumar-PV, and Ahmad-Dey. Cluster validations apply internal and relative criteria. The internal criteria include silhouette index and shadow values. The relative criterium applies bootstrap procedure producing a heatmap with a flexible reordering matrix algorithm such as ward, complete, or centroid linkages. The cluster result can be plotted in a marked barplot or pca biplot.

Copy Link

Version

Install

install.packages('kmed')

Monthly Downloads

330

Version

0.3.0

License

GPL-3

Maintainer

Weksi Budiaji

Last Published

June 14th, 2019

Functions in kmed (0.3.0)

distmix

Distances for mixed variables data set
matching

A pair distance for binary/ categorical variables
inckmed

Increasing number of clusters in k-medoids algorithm
heart

Heart Disease data set
stepkmed

Step k-medoid algorithm from Yu et al.
globalfood

Global food security index
fastkmed

Simple and fast k-medoid algorithm
rankkmed

Rank k-medoid algorithm
shadow

Centroid shadow value (CSV) index of each cluster based on medoid
pcabiplot

Biplot of a PCA object
silhoutte

Silhoutte index of each cluster
sil

Silhouette index and plot
clustboot

Bootstrap replications for clustering alorithm
clust5

5-clustered data set
distNumeric

A pair distance for numerical variables
clust4

4-clustered data set
barplotnum

Barplot of each cluster for numerical variables data set
cooccur

Co-occurrence distance for binary/ categorical variables data
csv

Centroid shadow value (CSV) index and plot
clustheatmap

Consensus matrix heatmap from A consensus matrix
consensusmatrix

Consensus matrix from A matrix of bootstrap replicates