CLUSTF

Clustering by the FFT (Farthest First Traversal) algorithm

   LAB = CLUSTF(A,K.MSIZE)
   LAB = A*CLUSTF(K.MSIZE)

Input
A	Feature based dataset with M objects.
K	Vector with desired numbers of clusters, default sampling of [2:M]
MSIZE	Number of objects (M) above which the dataset is preclustered by
CLUSTM,	reducing it to MSIZE objects. Default MSIZE = 5000. Use
MSIZE	= inf to avoid perclustering.

Output
LAB	M*N array with the results of the multi-level clusterings for the
M	objects. The columns refer to the N clusterings. They yield for  the objects the prototype indices of the clusters they belong to.

Description

The dataset A with M rows (objects) is clustered by the Farthest First  Traversal algorithm, initialised by the object nearest to the mean of A.  In every step a next cluster prototype is defined as the not yet selected  object in A that is farthest to the existing set of prototypes.  Objects are clustered by assigning them to the nearest prototype. The  dataset A can be either a double array or a PRTools dataset. Its labels  are negelected.

In LAB for all objects the index in A of the nearest cluster prototype is  returned. If K is a vector LAB has length(K) columns, returning a  multilevel clustering.

If M > MSIZE the dataset A is preclustered by PRECLUST using CLUSTM.  Unless specific values of K < 100 are needed it is recommended for fast  processing to use K = []. Speed may be further increased by using smaller  values of MSIZE, e.g. MSIZE = 500;

Example(s)

 randreset;                     % take care of reproducability
 data = gendatclust1(20000);    % generate 20000 objects in 10 clusters
                                % Run Mean Shift clustering
 lab = clustf(data,[2 5 10 18 30 50 100],2000);
                                % Show scatterplot for 10 clusters
 figure; scatn(lab(:,3),data,'FFT'); 
 figure; clusteval(lab,data);   % Evaluation by active learning

See also

datasets, mappings, cluste, clusth, clustk, clustkh, clustm, clusts, clustr, clusteval, clustcerr, clustc, clustnum, preclust,