Summary
International Symposium on Nonlinear Theory and its Applications
2005
Session Number:1-3-3
Session:
Number:1-3-3-4
Application of growing self-organizing map to small-world networking
Takeshi Ehara, Toshimichi Saito,
pp.258-261
Publication Date:2005/10/18
Online ISSN:2188-5079
DOI:10.34385/proc.40.1-3-3-4
PDF download (71.6KB)
Summary:
This paper studies a novel application of growing self-organizing maps to networking. In our algorithm nodes for the networking are applied successively as input data. Adapting to the input, the map can grow and can change the topology. Performing basic numerical experiments, we have confirmed that our algorithm can generate small-world like networks characterized by relatively small average path length and large clustering coefficient. These results provide basic information to develop interesting applications of unsupervised learning algorithms.