index.G1d {symbolicDA}

Calinski-Harabasz pseudo F-statistic based on distance matrix

Description

Calculates Calinski-Harabasz pseudo F-statistic based on distance matrix

Usage

index.G1d (d,cl)

Arguments

d

distance matrix (see dist_SDA)

cl

a vector of integers indicating the cluster to which each object is allocated

Details

See file ../doc/indexG1d_details.pdf for further details

Value

value of Calinski-Harabasz pseudo F-statistic based on distance matrix

Author(s)

Andrzej Dudek [e-mail], Justyna Wilk [e-mail] Department of Econometrics and Computer Science, Wroclaw University of Economics, Poland http://keii.ue.wroc.pl/symbolicDA/

References

Calinski, T., Harabasz, J. (1974), A dendrite method for cluster analysis, "Communications in Statistics", vol. 3, 1-27. Available at: doi:/10.1080/03610927408827101.

Everitt, B.S., Landau, E., Leese, M. (2001), Cluster analysis, Arnold, London, p. 103. ISBN 9780340761199.

Gordon, A.D. (1999), Classification, Chapman & Hall/CRC, London, p. 62. ISBN 9781584880134.

Milligan, G.W., Cooper, M.C. (1985), An examination of procedures of determining the number of cluster in a data set, "Psychometrika", vol. 50, no. 2, 159-179. Available at: doi:10.1007/BF02294245.

Diday, E., Noirhomme-Fraiture, M. (eds.) (2008), Symbolic Data Analysis with SODAS Software, John Wiley & Sons, Chichester, pp. 236-262.

Dudek, A. (2007), Cluster Quality Indexes for Symbolic Classification. An Examination, In: H.H.-J. Lenz, R. Decker (Eds.), Advances in Data Analysis, Springer-Verlag, Berlin, pp. 31-38. Available at: doi:10.1007/978-3-540-70981-7_4.

See Also

DClust, SClust; index.G2, index.G3, index.S, index.H,index.KL,index.Gap, index.DB in clusterSim library

Examples

# LONG RUNNING - UNCOMMENT TO RUN
# Example 1
#library(stats)
#data("cars",package="symbolicDA")
#x<-cars
#d<-dist_SDA(x, type="U_2")
#wynik<-hclust(d, method="ward", members=NULL)
#clusters<-cutree(wynik, 4)
#G1d<-index.G1d(d, clusters)
#print(G1d)

# Example 2


#data("cars",package="symbolicDA")
#md <- dist_SDA(cars, type="U_3", gamma=0.5, power=2)
# nc - number_of_clusters
#min_nc=2
#max_nc=10
#res <- array(0,c(max_nc-min_nc+1,2))
#res[,1] <- min_nc:max_nc
#clusters <- NULL
#for (nc in min_nc:max_nc)
#{
#cl2 <- pam(md, nc, diss=TRUE)
#res[nc-min_nc+1,2] <- G1d <- index.G1d(md,cl2$clustering)   
#clusters <- rbind(clusters, cl2$clustering)
#}
#print(paste("max G1d for",(min_nc:max_nc)[which.max(res[,2])],"clusters=",max(res[,2])))
#print("clustering for max G1d")
#print(clusters[which.max(res[,2]),])
#write.table(res,file="G1d_res.csv",sep=";",dec=",",row.names=TRUE,col.names=FALSE)
#plot(res, type="p", pch=0, xlab="Number of clusters", ylab="G1d", xaxt="n")
#axis(1, c(min_nc:max_nc))