Clasificador difuso neuronal aplicado a casos de datos sintéticos

Jose Gerardo Chacón Rangel, Anderson Smith Florez Fuentes, Johel Enrique Rodriguez Fernandez

Resumen


RESUMEN En este artículo se presenta el desarrollo de un sistema computacional difuso neuronal que permite clasificar casos de datos sintéticos a través patrones con solapamiento controlado. Se construyó una serie de modelos neuronales con lógica difusa y redes neurales que fueron analizados utilizando diferentes porcentajes de solapamiento. En función de los resultados obtenidos, se seleccionó el mejor modelo para clasificar los patrones de acuerdo con criterios apropiados de desempeño como error permitido y tiempo de entrenamiento. Se obtuvo un modelo capaz de identificar un tipo de clase, que tiende a minimizar los errores de clasificación. El modelo difuso neuronal de este tipo puede ayudar a especialistas de diferentes disciplinas a diagnosticar con un mínimo de error, cuando los datos presentan rasgos con patrones solapados. ABSTRACT This article presents the development of a computational system that allows diffuse neuronal classify cases of synthetic data through patterns overlap with controlled. They built a series of models with neural fuzzy logic and neural networks that were analyzed using different percentages of overlap. Depending on the results obtained, was selected the best model to classify the patterns in accordance with appropriate criteria for performance as permissible and training time. We obtained a model able to identify a type of class, which tends to minimize the errors of classification. The diffuse neuronal model of this type can help specialists from different disciplines to diagnose with a minimum of error, when data are traits with overlapping patterns.

Texto completo:

PDF

Referencias


Bishop, C. (1994). Neural Networks and their applications. Rev.Sci.Instrum., 65(6), 1803-1832.

Duch, W.; Adarnzak, R.; Grabczewski, K. (2001). A new methodology of extraction, optimization and application of crisp and fuzzy logical rules. IEEE trans. Neural Networks, 12, 277–306.

Duch, W.; Setiono, R.; Zurada, J. (2004). Computacional Methods for Rule-Based Data Understanding. Proceedings of the IEEE, 92(5), 771-805.

Hayashi, Y. (1994). Neural expert system using fuzzy teaching input and its application to medical diagnossis, Information scieces applications, 1, 47-58.

Hayashi, Y.; Imura, A.; Yoshida, K. (1990). Fuzzy neural expert system and its application to medical diagnosis, 8th International Congress on Cybernetics and Systems, New York City, 54-61.

Ma, L.; Chen, H.; Tian, Z.; He, W. (1999). Monitoring the particle size in CFB using fuzzy neural network. United States: American Society of Mechanical Engineers, New York, NY (US).

Mitra, S.; Hayashi, Y. (2000). Neuro-fuzzy rule generation: Survey in soft computing framework. IEEE Transactions on Neural Networks, 11, 748-768.

Pal, S.; Mitra, S. (1992). Multilayer perceptron, fuzzy sets, and classification. IEEE Transactions on Neural Networks, 3(5).

Pal, S.; Mitra, S. (1999). Neuro-Fuzzy Pattern Methods in soft Computing, Indian Statistical Institute Calcuta. Editorial Wiley and Sons, IN C.

Ramirez, C.; Vladimirova, T.A. (1995). Fuzzy neural network for fuzzy classification En IEEE International Conference on Systems, Man and Cybernetics. Vancouver, Canada, 322-327.




Licencia de Creative Commons
Este obra está bajo una licencia de Creative Commons Reconocimiento-NoComercial-SinObraDerivada 4.0 Internacional.

 

ISSN (Impreso): 2216-0353  y ISSN (En Línea): 2216-0388