Home > hit > clustering > hit_single_linkage.m

hit_single_linkage

PURPOSE ^

HIT_SINGLE_LINKAGE Single-linkage clustering algorithm.

SYNOPSIS ^

function [centers,cost,inl,class,etime,spec] = hit_single_linkage(V,OPT)

DESCRIPTION ^

HIT_SINGLE_LINKAGE Single-linkage clustering algorithm.

 -------------------------------------------------------------------------
 DESCRIPTION
 -------------------------------------------------------------------------
 [centers,cost,inl,class,etime,spec] = hit_single_linkage(V,OPT)

 -------------------------------------------------------------------------
 INPUT
 -------------------------------------------------------------------------
 V(i,:): i-th point to be clustered. Points are ROW vectors
 opt: structure of specific parameters
   MANDATORY FIELDS
   opt.guess: guessed minimal distance between clusters
   OPTIONAL FIELDS
   opt.plot_steps: 'Y' 'N' (set by default to 'N' if absent). It specify
   if the algorithm has to stop at each iteration and plot the clusters
   found. The algorithm will continue after that a key is pressed.

 -------------------------------------------------------------------------
 OUTPUT
 -------------------------------------------------------------------------
 centers(i,:): center of the i-th cluster (is a ROW vector).
 cost: value of the clustering cost functional.
 inl: indexes of datapoints that are inliers after clustering (i.e.
 not discarded by the clustering algorithm). Note that single-linkage
 performs outlier detection.
 class(i): classification of the i-th inlier. class(i)=j means
 that the i-th inlier belongs to the j-th cluster.
 etime: elapsed time for the execution of single-linkage.
 spec: structure with special outputs
   spec.costs(i) is the costs of the i-th cluster found.

CROSS-REFERENCE INFORMATION ^

This function calls: This function is called by:
Generated on Thu 01-Dec-2005 10:54:38 by m2html © 2003