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
Cet article propose un algorithme de carte auto-organisé efficace basé sur un point de référence et des filtres. Une stratégie appelée Reference Point SOM (RPSOM) est proposée pour améliorer le temps d'exécution du SOM au moyen d'un filtrage à deux seuils. T1 et à la T2. Nous utilisons un seuil, T1, pour définir le paramètre de limite de recherche utilisé pour rechercher l'unité la mieux adaptée (BMU) par rapport aux vecteurs d'entrée. L'autre seuil, T2, est utilisé comme limite de recherche à l'intérieur de laquelle le BMU trouve ses voisins. L'algorithme proposé réduit la complexité temporelle de O(n2) À O(n) pour trouver les neurones initiaux par rapport à l'algorithme proposé par Su et al. [16] . Le RPSOM réduit considérablement la complexité temporelle, en particulier dans le calcul d'un grand ensemble de données. À partir des résultats expérimentaux, nous constatons qu’il est préférable de construire une bonne carte initiale puis d’utiliser l’apprentissage non supervisé pour effectuer de petits ajustements ultérieurs.
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.
Copier
Shu-Ling SHIEH, I-En LIAO, Kuo-Feng HWANG, Heng-Yu CHEN, "An Efficient Initialization Scheme for SOM Algorithm Based on Reference Point and Filters" in IEICE TRANSACTIONS on Information,
vol. E92-D, no. 3, pp. 422-432, March 2009, doi: 10.1587/transinf.E92.D.422.
Abstract: This paper proposes an efficient self-organizing map algorithm based on reference point and filters. A strategy called Reference Point SOM (RPSOM) is proposed to improve SOM execution time by means of filtering with two thresholds T1 and T2. We use one threshold, T1, to define the search boundary parameter used to search for the Best-Matching Unit (BMU) with respect to input vectors. The other threshold, T2, is used as the search boundary within which the BMU finds its neighbors. The proposed algorithm reduces the time complexity from O(n2) to O(n) in finding the initial neurons as compared to the algorithm proposed by Su et al. [16] . The RPSOM dramatically reduces the time complexity, especially in the computation of large data set. From the experimental results, we find that it is better to construct a good initial map and then to use the unsupervised learning to make small subsequent adjustments.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E92.D.422/_p
Copier
@ARTICLE{e92-d_3_422,
author={Shu-Ling SHIEH, I-En LIAO, Kuo-Feng HWANG, Heng-Yu CHEN, },
journal={IEICE TRANSACTIONS on Information},
title={An Efficient Initialization Scheme for SOM Algorithm Based on Reference Point and Filters},
year={2009},
volume={E92-D},
number={3},
pages={422-432},
abstract={This paper proposes an efficient self-organizing map algorithm based on reference point and filters. A strategy called Reference Point SOM (RPSOM) is proposed to improve SOM execution time by means of filtering with two thresholds T1 and T2. We use one threshold, T1, to define the search boundary parameter used to search for the Best-Matching Unit (BMU) with respect to input vectors. The other threshold, T2, is used as the search boundary within which the BMU finds its neighbors. The proposed algorithm reduces the time complexity from O(n2) to O(n) in finding the initial neurons as compared to the algorithm proposed by Su et al. [16] . The RPSOM dramatically reduces the time complexity, especially in the computation of large data set. From the experimental results, we find that it is better to construct a good initial map and then to use the unsupervised learning to make small subsequent adjustments.},
keywords={},
doi={10.1587/transinf.E92.D.422},
ISSN={1745-1361},
month={March},}
Copier
TY - JOUR
TI - An Efficient Initialization Scheme for SOM Algorithm Based on Reference Point and Filters
T2 - IEICE TRANSACTIONS on Information
SP - 422
EP - 432
AU - Shu-Ling SHIEH
AU - I-En LIAO
AU - Kuo-Feng HWANG
AU - Heng-Yu CHEN
PY - 2009
DO - 10.1587/transinf.E92.D.422
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E92-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2009
AB - This paper proposes an efficient self-organizing map algorithm based on reference point and filters. A strategy called Reference Point SOM (RPSOM) is proposed to improve SOM execution time by means of filtering with two thresholds T1 and T2. We use one threshold, T1, to define the search boundary parameter used to search for the Best-Matching Unit (BMU) with respect to input vectors. The other threshold, T2, is used as the search boundary within which the BMU finds its neighbors. The proposed algorithm reduces the time complexity from O(n2) to O(n) in finding the initial neurons as compared to the algorithm proposed by Su et al. [16] . The RPSOM dramatically reduces the time complexity, especially in the computation of large data set. From the experimental results, we find that it is better to construct a good initial map and then to use the unsupervised learning to make small subsequent adjustments.
ER -