ClusterTools Contents

ClusterTools User Guide

dclust2proto

DCLUST2PROTO

Find prototypes from given clustering of dissimilarity data

    LAB_OUT = DCLUST2PROTO(D,LABC)
    LAB_OUT = LABC*DCLUST2PROTO(D)

Input
 LABC Index array, size [M,K], indices of cluster prototypes for M objects in K clusterings.
 D M x M PRTools dissimilarity matrix or double array

Output
 LAB_OUT Vector with set of M indices pointing for every object in A to the centre of its cluster as derived from D.

Description

By this routine cluster labels of a dataset D are replaced by the indices  of the centres of these clusters. A cluster centre is the object for  which the maximum distance to the other objects in the cluster is minimized.

In case the cluster labels LAB_IN are not given D should be a PRTools  dataset and its labels are interpreted as the cluster labels.

See also

mappings, exemplar, dcluste, dclustm, dclusth, dclustk,

ClusterTools Contents

ClusterTools User Guide

This file has been automatically generated. If badly readable, use the help-command in Matlab.