EarthRef.org Digital Archive (ERDA) -- KKAnalysis - A software for unsupervised classification by SOM and Cluster Analysis

Description
KKAnalysis is a collection of methods for unsupervised classification and clustering. The software has been developed in MATLAB exploiting routines of the SOM Toolbox 2 for MATLAB (http://www.cis.hut.fi /projects/somtoolbox/). Besides functions related to the SOM, we have added K-Means, Fuzzy C-Means and adaptive determinant clustering (the latter taken from Späth, 1983). All clustering schemes can be applied to the map created by the SOM routines and to the original patterns as well. KKAnalysis provides both numerical and graphical output. In order to render the program user-friendly a specific GUI has been created where every parameter is easily configurable by a click on the appropriate control. KKAnalysis has been released for Microsoft Windows systems (tested on XP and Vista) as well as Linux based systems (tested on Ubuntu 9.04).
This contribution consists of KKAnalysis executable files (Windows and Linux versions), the Example Dataset, the Extended Documentation and the video-tutorial.