HINoV.Symbolic {clusterSim}

Modification of Carmone, Kara & Maxwell Heuristic Identification of Noisy Variables (HINoV) method for symbolic interval data

Description

Modification of Heuristic Identification of Noisy Variables (HINoV) method for symbolic interval data

Usage

HINoV.Symbolic(x, u=NULL, distance="H", method = "pam", 
	Index = "cRAND")

Arguments

x

symbolic interval data: a 3-dimensional table, first dimension represents object number, second dimension - variable number, and third dimension contains lower- and upper-bounds of intervals

u

number of clusters

distance

"M" - minimal distance between all vertices of hyper-cubes defined by symbolic interval variables; "H" - Hausdorff distance; "S" - sum of squares of distance between all vertices of hyper-cubes defined by symbolic interval variables

method

clustering method: "single", "ward.D", "ward.D2", "complete", "average", "mcquitty", "median", "centroid", "pam" (default)

Index

"cRAND" - corrected Rand index (default); "RAND" - Rand index

Details

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

Value

parim

m x m symmetric matrix (m - number of variables). Matrix contains pairwise corrected Rand (Rand) indices for partitions formed by the j-th variable with partitions formed by the l-th variable

topri

sum of rows of parim

stopri

ranked values of topri in decreasing order

Author(s)

Marek Walesiak [e-mail], Andrzej Dudek [e-mail]

Department of Econometrics and Computer Science, University of Economics, Wroclaw, Poland http://keii.ue.wroc.pl/clusterSim/

References

Carmone, F.J., Kara, A., Maxwell, S. (1999), HINoV: a new method to improve market segment definition by identifying noisy variables, "Journal of Marketing Research", November, vol. 36, 501-509.

Hubert, L.J., Arabie, P. (1985), Comparing partitions, "Journal of Classification", no. 1, 193-218. Available at: doi:10.1007/BF01908075.

Rand, W.M. (1971), Objective criteria for the evaluation of clustering methods, "Journal of the American Statistical Association", no. 336, 846-850. Available at: doi:10.1080/01621459.1971.10482356.

Walesiak, M., Dudek, A. (2008), Identification of noisy variables for nonmetric and symbolic data in cluster analysis, In: C. Preisach, H. Burkhardt, L. Schmidt-Thieme, R. Decker (Eds.), Data analysis, machine learning and applications, Springer-Verlag, Berlin, Heidelberg, 85-92. Available at: http://keii.ue.wroc.pl/pracownicy/mw/2010_Walesiak_Dudek_Springer.PDF

See Also

hclust, kmeans, cluster.Sim

Examples

library(clusterSim)
data(data_symbolic)
r<- HINoV.Symbolic(data_symbolic, u=5)
print(r$stopri)
plot(r$stopri[,2], xlab="Variable number", ylab="topri",
xaxt="n", type="b")
axis(1,at=c(1:max(r$stopri[,1])),labels=r$stopri[,1])

#symbolic data from .csv file
#library(clusterSim)
#dsym<-as.matrix(read.csv2(file="csv/symbolic.csv"))
#dim(dsym)<-c(dim(dsym)[1],dim(dsym)[2]%/%2,2)          
#r<- HINoV.Symbolic(dsym, u=5)
#print(r$stopri)
#plot(r$stopri[,2], xlab="Variable number", ylab="topri",
#xaxt="n", type="b")
#axis(1,at=c(1:max(r$stopri[,1])),labels=r$stopri[,1])