Cluster centers of a classification

Function clustcentroid calculates the centroid (multivariate average) coordinates of a classification. Function clustmedoid determines the medoid (object whose average dissimilarity to all the other objects is minimal) for each cluster in the classification.

Usage

clustcentroid(x, y, m = 1)
clustmedoid(x, y, m = 1)

Arguments

x: Community data, a site-by-species data frame. In function clustmedoid, x can alternatively be an object of class dist (otherwise, the dissimilarity measure is assumed to be the Euclidean distance).
y: It can be (a) A vector indicating the cluster that each object in x belongs to; (b) a fuzzy/hard site-by-group matrix of membership values; (c) an object of class vegclust or vegclass
m: Fuzziness exponent, only effective when y is a fuzzy membership matrix.

Value

Function clustcentroid returns a group-by-species matrix containing species average abundance values (i.e. the coordinates of each cluster centroid). Function clustmedoid returns a vector of indices (medoids).

Note

In order to assign new plot record data into a predefined set of classes, one should use functions as.vegclust and vegclass instead.

Author

Miquel De Cáceres, CREAF

Examples

## Loads stats
library(stats)
  
## Loads data
data(wetland)
  
## This equals the chord transformation 
## (see also \code{\link{decostand}} in package 'vegan')
wetland.chord = as.data.frame(sweep(as.matrix(wetland), 1,
                              sqrt(rowSums(as.matrix(wetland)^2)), "/"))

## Performs a K-means clustering
wetland.km = kmeans(wetland.chord, centers=3, nstart=10)

## Gets the coordinates corresponding to the centroids of KM clusters
clustcentroid(wetland.chord, y=wetland.km$cluster)
#>       Abefic     Merhed     Alyvag     Pancam     Abemos     Melcor     Ludoct
#> 1 0.00000000 0.00000000 0.04696958 0.07606533 0.01558699 0.15317754 0.00000000
#> 2 0.00000000 0.00000000 0.00000000 0.02725749 0.00000000 0.06157316 0.00000000
#> 3 0.01825742 0.06770211 0.00000000 0.35635911 0.01458650 0.27504998 0.01889822
#>       Eupvac     Echpas     Passcr       Poa2     Carhal     Dendio     Casobt
#> 1 0.05023873 0.06646064 0.00000000 0.01041893 0.06911470 0.00000000 0.00000000
#> 2 0.03879374 0.04762643 0.00000000 0.00000000 0.03679444 0.01572125 0.01631471
#> 3 0.20498936 0.10174214 0.02581989 0.00000000 0.00000000 0.00000000 0.00000000
#>       Aesind     Cyprot     Ipocop     Cynarc     Walind    Sessp.     Phynod
#> 1 0.07475403 0.08126971 0.01262691 0.01262691 0.06377158 0.1041481 0.71534308
#> 2 0.00000000 0.00000000 0.00000000 0.00000000 0.06827998 0.1437923 0.00000000
#> 3 0.01889822 0.17597447 0.03974967 0.43238581 0.07754611 0.1569788 0.06361634
#>      Echell     Helind     Ipoaqu    Orysp.    Elesp.     Psespi   Ludads
#> 1 0.1191200 0.17218890 0.04775048 0.1720463 0.1377765 0.02691262 0.000000
#> 2 0.0491431 0.12531919 0.05532759 0.6491987 0.1535492 0.19245622 0.204177
#> 3 0.1120500 0.01889822 0.03657589 0.0400000 0.0000000 0.00000000 0.000000
#>       Polatt       Poa1     Helcri     Physp.     Goopur
#> 1 0.00000000 0.01282895 0.01282895 0.00000000 0.00000000
#> 2 0.09678166 0.00000000 0.00000000 0.01153624 0.01153624
#> 3 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000

## Gets the object indices corresponding to the medoids of KM clusters
clustmedoid(wetland.chord, y=wetland.km$cluster)
#> 12  2 24 
#> 23 11 34