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
Dans cet article, le modèle LVQ (Learning Vector Quantization) et ses variantes sont considérés comme des outils de clustering permettant de discriminer les événements sismiques naturels (tremblements de terre) des événements artificiels (explosions nucléaires). L'étude est basée sur les six caractéristiques spectrales des spectres d'onde P calculées à partir des enregistrements télésismiques de courte période. Les modèles LVQ conventionnels proposés par Kohenen ainsi que les modèles Fuzzy LVQ (FLVQ) proposés par Sakuraba et Bezdek sont tous testés sur un ensemble de 26 tremblements de terre et 24 explosions nucléaires en utilisant le laisser-un-dehors stratégie de test. Les principaux résultats expérimentaux ont montré que les formes, le nombre ainsi que les chevauchements des clusters jouent un rôle important dans la classification sismique. Les résultats ont également montré comment un partitionnement inapproprié de l’espace des fonctionnalités affaiblirait fortement les phases de clustering et de reconnaissance. Pour améliorer les résultats numériques, un nouvel algorithme FLVQ combiné est utilisé dans cet article. L'algorithme est composé de deux sous-algorithmes imbriqués. Le sous-algorithme interne tente de générer un partitionnement flou bien défini avec les vecteurs de référence flous dans l'espace des fonctionnalités. Pour atteindre cet objectif, une fonction de coût est définie en fonction du nombre, des formes mais aussi des recouvrements des vecteurs de référence flous. La règle de mise à jour tente de minimiser cette fonction de coût dans un algorithme d'apprentissage pas à pas. D'autre part, le sous-algorithme externe tente de trouver une valeur optimale pour le nombre de clusters, à chaque étape. Pour cette optimisation dans la boucle externe, nous avons utilisé deux critères différents. Dans le premier critère, le nouveau "entropie floue" est utilisé tandis que dans le deuxième critère, un indice de performance est employé en généralisant la formule de Huntsberger pour le taux d'apprentissage, en utilisant la notion de distance floue. Les résultats expérimentaux du nouveau modèle montrent une amélioration prometteuse du taux d'erreur, un temps de convergence acceptable, ainsi qu'une plus grande flexibilité dans la prise de décision concernant les limites.
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
Payam NASSERY, Karim FAEZ, "Seismic Events Discrimination Using a New FLVQ Clustering Model" in IEICE TRANSACTIONS on Information,
vol. E83-D, no. 7, pp. 1533-1539, July 2000, doi: .
Abstract: In this paper, the LVQ (Learning Vector Quantization) model and its variants are regarded as the clustering tools to discriminate the natural seismic events (earthquakes) from the artificial ones (nuclear explosions). The study is based on the six spectral features of the P-wave spectra computed from the short period teleseismic recordings. The conventional LVQ proposed by Kohenen and also the Fuzzy LVQ (FLVQ) models proposed by Sakuraba and Bezdek are all tested on a set of 26 earthquakes and 24 nuclear explosions using the leave-one-out testing strategy. The primary experimental results have shown that the shapes, the number and also the overlaps of the clusters play an important role in seismic classification. The results also showed how an improper feature space partitioning would strongly weaken both the clustering and recognition phases. To improve the numerical results, a new combined FLVQ algorithm is employed in this paper. The algorithm is composed of two nested sub-algorithms. The inner sub-algorithm tries to generate a well-defined fuzzy partitioning with the fuzzy reference vectors in the feature space. To achieve this goal, a cost function is defined as a function of the number, the shapes and also the overlaps of the fuzzy reference vectors. The update rule tries to minimize this cost function in a stepwise learning algorithm. On the other hand, the outer sub-algorithm tries to find an optimum value for the number of the clusters, in each step. For this optimization in the outer loop, we have used two different criteria. In the first criterion, the newly defined "fuzzy entropy" is used while in the second criterion, a performance index is employed by generalizing the Huntsberger formula for the learning rate, using the concept of fuzzy distance. The experimental results of the new model show a promising improvement in the error rate, an acceptable convergence time, and also more flexibility in boundary decision making.
URL: https://global.ieice.org/en_transactions/information/10.1587/e83-d_7_1533/_p
Copier
@ARTICLE{e83-d_7_1533,
author={Payam NASSERY, Karim FAEZ, },
journal={IEICE TRANSACTIONS on Information},
title={Seismic Events Discrimination Using a New FLVQ Clustering Model},
year={2000},
volume={E83-D},
number={7},
pages={1533-1539},
abstract={In this paper, the LVQ (Learning Vector Quantization) model and its variants are regarded as the clustering tools to discriminate the natural seismic events (earthquakes) from the artificial ones (nuclear explosions). The study is based on the six spectral features of the P-wave spectra computed from the short period teleseismic recordings. The conventional LVQ proposed by Kohenen and also the Fuzzy LVQ (FLVQ) models proposed by Sakuraba and Bezdek are all tested on a set of 26 earthquakes and 24 nuclear explosions using the leave-one-out testing strategy. The primary experimental results have shown that the shapes, the number and also the overlaps of the clusters play an important role in seismic classification. The results also showed how an improper feature space partitioning would strongly weaken both the clustering and recognition phases. To improve the numerical results, a new combined FLVQ algorithm is employed in this paper. The algorithm is composed of two nested sub-algorithms. The inner sub-algorithm tries to generate a well-defined fuzzy partitioning with the fuzzy reference vectors in the feature space. To achieve this goal, a cost function is defined as a function of the number, the shapes and also the overlaps of the fuzzy reference vectors. The update rule tries to minimize this cost function in a stepwise learning algorithm. On the other hand, the outer sub-algorithm tries to find an optimum value for the number of the clusters, in each step. For this optimization in the outer loop, we have used two different criteria. In the first criterion, the newly defined "fuzzy entropy" is used while in the second criterion, a performance index is employed by generalizing the Huntsberger formula for the learning rate, using the concept of fuzzy distance. The experimental results of the new model show a promising improvement in the error rate, an acceptable convergence time, and also more flexibility in boundary decision making.},
keywords={},
doi={},
ISSN={},
month={July},}
Copier
TY - JOUR
TI - Seismic Events Discrimination Using a New FLVQ Clustering Model
T2 - IEICE TRANSACTIONS on Information
SP - 1533
EP - 1539
AU - Payam NASSERY
AU - Karim FAEZ
PY - 2000
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E83-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2000
AB - In this paper, the LVQ (Learning Vector Quantization) model and its variants are regarded as the clustering tools to discriminate the natural seismic events (earthquakes) from the artificial ones (nuclear explosions). The study is based on the six spectral features of the P-wave spectra computed from the short period teleseismic recordings. The conventional LVQ proposed by Kohenen and also the Fuzzy LVQ (FLVQ) models proposed by Sakuraba and Bezdek are all tested on a set of 26 earthquakes and 24 nuclear explosions using the leave-one-out testing strategy. The primary experimental results have shown that the shapes, the number and also the overlaps of the clusters play an important role in seismic classification. The results also showed how an improper feature space partitioning would strongly weaken both the clustering and recognition phases. To improve the numerical results, a new combined FLVQ algorithm is employed in this paper. The algorithm is composed of two nested sub-algorithms. The inner sub-algorithm tries to generate a well-defined fuzzy partitioning with the fuzzy reference vectors in the feature space. To achieve this goal, a cost function is defined as a function of the number, the shapes and also the overlaps of the fuzzy reference vectors. The update rule tries to minimize this cost function in a stepwise learning algorithm. On the other hand, the outer sub-algorithm tries to find an optimum value for the number of the clusters, in each step. For this optimization in the outer loop, we have used two different criteria. In the first criterion, the newly defined "fuzzy entropy" is used while in the second criterion, a performance index is employed by generalizing the Huntsberger formula for the learning rate, using the concept of fuzzy distance. The experimental results of the new model show a promising improvement in the error rate, an acceptable convergence time, and also more flexibility in boundary decision making.
ER -