RSDA (version 2.0.8)

sym.hclust: Symbolic Hierarchical Clustering

Description

This function allows us to execute a symbolic hierarchical clustering to interval variables.

Usage

sym.hclust(sym.data, distance = c('hausdorff', 'centers'), p = 2,
method = c('ward.D2', 'single', 'complete', 'average', 'mcquitty',
           'median', 'centroid'), members = NULL)

Arguments

sym.data

The symbolic data table.

distance

The distance to be use.

p

The p in the Hausdorff distance : $$d(w_{u_1},w_{u_2}) = \left( \sum_{j=1}^m \Phi_j(w_{u_1},w_{u_2})^p \right)^{1/p}$$

method

The method to be use, like in hclust R function.

members

Like in hclust R function.

Value

Return a dendogram plot structure.

References

Carvalho F., Souza R.,Chavent M., and Lechevallier Y. (2006) Adaptive Hausdorff distances and dynamic clustering of symbolic interval data. Pattern Recognition Letters Volume 27, Issue 3, February 2006, Pages 167-179

Rodriguez, O. (2000). Classification et Modeles Lineaires en Analyse des Donnees Symboliques. Ph.D. Thesis, Paris IX-Dauphine University.

Examples

# NOT RUN {
data(oils)
sh<-sym.hclust(oils)
plot(sh)
sh<-sym.hclust(oils,'centers')
plot(sh)
# }