The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Recentemente, mapas de características foram aplicados a vários domínios de problemas. O sucesso de algumas dessas aplicações depende criticamente de os mapas de características serem ordenados topologicamente. Várias abordagens diferentes foram propostas para melhorar o algoritmo convencional de mapa de características auto-organizadas (SOM). No entanto, estas abordagens não garantem que um mapa de características topologicamente ordenado possa ser formado no final de uma simulação. Portanto, o procedimento de tentativa e erro ainda domina o procedimento de formação de mapas de características. Neste artigo, propomos um mecanismo de cura para reparar mapas de características que não estão bem ordenados topologicamente. O mapa curado é então ajustado pelo algoritmo SOM convencional com uma pequena taxa de aprendizagem e um conjunto de vizinhança de pequeno tamanho, de modo a melhorar a precisão do mapa. Dois conjuntos de dados foram testados para ilustrar o desempenho do método proposto.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Mu-Chun SU, Chien-Hsing CHOU, Hsiao-Te CHANG, "A Healing Mechanism to Improve the Topological Preserving Property of Feature Maps" in IEICE TRANSACTIONS on Information,
vol. E85-D, no. 4, pp. 735-743, April 2002, doi: .
Abstract: Recently, feature maps have been applied to various problem domains. The success of some of these applications critically depends on whether feature maps are topologically ordered. Several different approaches have been proposed to improve the conventional self-organizing feature map (SOM) algorithm. However, these approaches do not guarantee that a topologically-ordered feature map can be formed at the end of a simulation. Therefore, the trial-and-error procedure still dominates the procedure of forming feature maps. In this paper, we propose a healing mechanism to repair feature maps that are not well topologically ordered. The healed map is then further fine-tuned by the conventional SOM algorithm with a small learning rate and a small-sized neighborhood set so as to improve the accuracy of the map. Two data sets were tested to illustrate the performance of the proposed method.
URL: https://global.ieice.org/en_transactions/information/10.1587/e85-d_4_735/_p
Copiar
@ARTICLE{e85-d_4_735,
author={Mu-Chun SU, Chien-Hsing CHOU, Hsiao-Te CHANG, },
journal={IEICE TRANSACTIONS on Information},
title={A Healing Mechanism to Improve the Topological Preserving Property of Feature Maps},
year={2002},
volume={E85-D},
number={4},
pages={735-743},
abstract={Recently, feature maps have been applied to various problem domains. The success of some of these applications critically depends on whether feature maps are topologically ordered. Several different approaches have been proposed to improve the conventional self-organizing feature map (SOM) algorithm. However, these approaches do not guarantee that a topologically-ordered feature map can be formed at the end of a simulation. Therefore, the trial-and-error procedure still dominates the procedure of forming feature maps. In this paper, we propose a healing mechanism to repair feature maps that are not well topologically ordered. The healed map is then further fine-tuned by the conventional SOM algorithm with a small learning rate and a small-sized neighborhood set so as to improve the accuracy of the map. Two data sets were tested to illustrate the performance of the proposed method.},
keywords={},
doi={},
ISSN={},
month={April},}
Copiar
TY - JOUR
TI - A Healing Mechanism to Improve the Topological Preserving Property of Feature Maps
T2 - IEICE TRANSACTIONS on Information
SP - 735
EP - 743
AU - Mu-Chun SU
AU - Chien-Hsing CHOU
AU - Hsiao-Te CHANG
PY - 2002
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E85-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2002
AB - Recently, feature maps have been applied to various problem domains. The success of some of these applications critically depends on whether feature maps are topologically ordered. Several different approaches have been proposed to improve the conventional self-organizing feature map (SOM) algorithm. However, these approaches do not guarantee that a topologically-ordered feature map can be formed at the end of a simulation. Therefore, the trial-and-error procedure still dominates the procedure of forming feature maps. In this paper, we propose a healing mechanism to repair feature maps that are not well topologically ordered. The healed map is then further fine-tuned by the conventional SOM algorithm with a small learning rate and a small-sized neighborhood set so as to improve the accuracy of the map. Two data sets were tested to illustrate the performance of the proposed method.
ER -