Hierarchical reclustering of a multilevel clustering
[LABOUT,DEN] = RECLUSTH(LABIN,K,TYPE)
[LABOUT,DEN] = LABIN*RECLUSTH(K,TYPE)
| LABIN|| Multilevel labeling, size [M,L], for M objects.|
| K|| Vector with desired numbers of clusters, default sampling of [2:M] |
| TYPE|| Linkage rule: 's', 'a', 'r' or 'c'. Alternatively 'single','average','central' or 'complete'. Default 'complete'.|
| LABOUT|| Index array, size [M,length(K)], indices of cluster prototypes. Columns refer to different clusterings and are ranked to increasing numbers of clusters.|
| DEN|| Dendrogram, see PLOTDG or DENDROGRAM |
This routine performs an agglomorative hierarchical clustering based on a distance matrix between all clusters on the highest level of LABIN. All clusterings are used to compute the distances based on histograms of cluster assignments of all objects in a cluster.
The quality of this routine may be improved considerably by using a very diverse set of input labelings LABIN.
datasets, mappings, dclusth, cluste, clustf, clustk, clusth, clustkh, clustm, reclustk, reclustn, clusteval, clustcerr, clustc, clustnum, plotdg,
|This file has been automatically generated. If badly readable, use the help-command in Matlab.|