Universität Bielefeld Electronic Collections animiertes Foto Universität Bielefeld

Access to the Document



Description of Input Patterns by Linear Mixtures of SOM Models

Kohonen, Teuvo



Download file

Abstract:
This paper introduces a novel way of analyzing input patterns presented to the Self-Organizing Map (SOM). Instead of identifying only the "winner," i.e., the model that matches best with the input, we determine the linear mixture of the models (reference vectors) of the SOM that approximates to the input vector best. It will be shown that if only nonnegative weights are allowed in this linear mixture, the expansion of the input pattern in terms of the models is very meaningful, contains only few terms, and provides a better insight into the input state than what the mere "winner" can give. If then the models fall into classes that are known a priori, the sums of the weights over each class can be interpreted as expressing the affiliation of the input with the due classes.


Keywords: least-squares fitting, linear mixture, self-organizing map
Institution: Faculty of Technology, Research Groups in Informatics
DDC classification: Data processing, computer science, computer systems

Suggested Citation:
Kohonen, Teuvo  (2007)  Description of Input Patterns by Linear Mixtures of SOM Models.


URL: http://biecoll.ub.uni-bielefeld.de/volltexte/2007/123



 Questions or comments: publikationsdienste.ub@uni-bielefeld.de
 Latest update: 15 Feb 2011
 Legal Notice
OPUS-Logo     OAI compliant      BU Logo
OAI-Logo