fpc (version 2.2-9)

cvnn: Cluster validation based on nearest neighbours

Description

Cluster validity index based on nearest neighbours as defined in Liu et al. (2013) with a correction explained in Halkidi et al. (2015).

Usage

cvnn(d=NULL,clusterings,k=5)

Arguments

d

dissimilarity matrix or dist-object.

clusterings

list of vectors of integers with length =nrow(d); indicating the cluster for each observation for several clusterings (list elements) to be compared.

k

integer. Number of nearest neighbours.

Value

List with components (see Liu et al. (2013), Halkidi et al. (2015) for details)

cvnnindex

vector of index values for the various clusterings, see Liu et al. (2013), the lower the better.

sep

vector of separation values.

comp

vector of compactness values.

References

Halkidi, M., Vazirgiannis, M. and Hennig, C. (2015) Method-independent indices for cluster validation. In C. Hennig, M. Meila, F. Murtagh, R. Rocci (eds.) Handbook of Cluster Analysis, CRC Press/Taylor & Francis, Boca Raton.

Liu, Y, Li, Z., Xiong, H., Gao, X., Wu, J. and Wu, S. (2013) Understanding and enhancement of internal clustering validation measures. IEEE Transactions on Cybernetics 43, 982-994.

Examples

# NOT RUN {
  options(digits=3)
  iriss <- as.matrix(iris[c(1:10,51:55,101:105),-5])
  irisc <- as.numeric(iris[c(1:10,51:55,101:105),5])
  print(cvnn(dist(iriss),list(irisc,rep(1:4,5))))
# }