Unexpected results of SOM learning and its detection

Kohonen's Self Organizing Map (SOM) involves neural networks, for which an algorithm learns the feature of input data through unsupervised, competitive neighborhood learning. In many cases of SOM learning, if the data make classes in input data space with similar density, similar shape, and sim...

Full description

Saved in:
Bibliographic Details
Published in2010 IEEE International Conference on Systems, Man and Cybernetics pp. 3569 - 3574
Main Authors Miyoshi, T, Nishii, Y
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2010
Subjects
Artificial intelligence
feature of data
Integrated circuits
learning
self organizing map
Online AccessGet full text

Cover

More Information
Summary:Kohonen's Self Organizing Map (SOM) involves neural networks, for which an algorithm learns the feature of input data through unsupervised, competitive neighborhood learning. In many cases of SOM learning, if the data make classes in input data space with similar density, similar shape, and similar size, corresponding classes in feature map also formed to similar shape and similar size. In the experiments, however, we found unexpected learning results, corresponding classes in feature map formed to different shape and different size one another. In this paper, we investigate what kind of learning data set, which feature of learning data causes unexpected results.
ISBN:1424465869
9781424465866
ISSN:1062-922X
2577-1655
DOI:10.1109/ICSMC.2010.5642344