The Kohonen and CP-ANN toolbox: A collection of MATLAB modules for Self Organizing Maps and Counterpropagation Artificial Neural Networks

摘要

Kohonen maps (or Self Organizing Maps, SOMs) and Counterpropagation Artificial Neural Networks are two of the most popular Neural Networks proposed in literature and are increasing in uses and applications related to several multivariate chemical issues, since they can handle both supervised and unsupervised problems. This work deals with the presentation of the Kohonen and CP-ANN toolbox, that is a collection of MATLAB modules freely available via the Internet (http://www.disat.unimib.it/chm) for the calculation of the quoted models. A graphical user interface (GUI), which allows an easy model calculation and analysis of results, is also provided. The toolbox features are presented by reproducing the classification of a real multivariate dataset. This work is not an attempt to summarize the general applications of Self Organizing Maps, but to inform chemometricians and practitioners who are not skilled programmers of the existence of a user-friendly Matlab toolbox to develop unsupervised and supervised SOM models.