clv.SD, clv.SDbw {clv}

SD, SDbw - Internal Measures

Description

Function computes SD and \textrm{S\_Dbw} validity indices.

Usage

clv.SD(scatt, dis, alfa)
clv.SDbw(scatt, dens)

Arguments

scatt

average scattering for cluster value computed using clv.Scatt function.

dis

total separation between clusters value computed using clv.Dis function.

dens

inter-cluster density value computed using clv.DensBw function.

alfa

weighting factor (normally equal to Dis(cmax) where cmax is the maximum number of input clusters).

Details

SD validity index is defined by equation:

SD = scatt*alfa + dis

where scatt means average scattering for clusters defined in clv.Scatt. \textrm{S\_Dbw} validity index is defined by equation:

\textrm{S\_Dbw} = scatt + dens

where dens is defined in clv.DensBw.

Value

As result of clv.SD function SD validity index is returned. As result of clv.SDbw function \textrm{S\_Dbw} validity index is returned.

Author(s)

Lukasz Nieweglowski

References

M. Haldiki, Y. Batistakis, M. Vazirgiannis On Clustering Validation Techniques, http://citeseer.ist.psu.edu/513619.html

See Also

clv.Scatt, clv.Dis and clv.DensBw

Examples

# load and prepare
library(clv)
data(iris)
iris.data <- iris[,1:4]

# cluster data
agnes.mod <- agnes(iris.data) # create cluster tree 
v.pred <- as.integer(cutree(agnes.mod,5)) # "cut" the tree 

# prepare proper input data for SD and S_Dbw indicies
scatt <- clv.Scatt(iris.data, v.pred)
dis <- clv.Dis(scatt$cluster.center)
dens.bw <- clv.DensBw(iris.data, v.pred, scatt)

# compute  SD and S_Dbw indicies
SD <- clv.SD(scatt$Scatt, dis, alfa=5) # alfa is equal to number of clusters 
SDbw <- clv.SDbw(scatt$Scatt, dens.bw)