SciELO - Scientific Electronic Library Online

 
vol.8 issue1An Adaptive Coherence-Replacement Protocol for Web Proxy Cache SystemsSpecification and Analysis of Real Time Systems in Type Theory author indexsubject indexsearch form
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • Have no similar articlesSimilars in SciELO

Share


Computación y Sistemas

On-line version ISSN 2007-9737Print version ISSN 1405-5546

Comp. y Sist. vol.8 n.1 Ciudad de México Jul./Sep. 2004

 

Feature Selection using Typical Testors applied to Estimation of Stellar Parameters

 

Selección de Características usando Testores Típicos aplicada a la Estimación de Parámetros Estelares

 

José Á. Santos, Ariel Carrasco and José F. Martínez

 

1 Instituto Nacional de Astrofísica Óptica y Electrónica (INAOE), México angellisse@hotmail.com ; ariel@inaoep.mx ; fmartinez@inaoep.mx

 

Article received on august 18, 2003
Accepted on august 05, 2004

 

Abstract

In this paper a comparative analysis of feature selection using typical testors applied on astronomical data, is presented. The comparison is based on the classification efficiency using typical testors as feature selection method against the classification efficiency using Ramirez (2001) method, which uses genetic algorithms. The well–known K–nearest neighbors rule (KNN) was used as classifier. The feature selection based on typical testors was modified to be applied on a prediction problem of a real valued function. The feature selection obtained with typical testors reduces the amount of features in approximately 50% and the classification error index is better than both using the original data and Ramirez's method.

Keywords: Feature Selection; Typical Testors; Logical Combinatorial Pattern Recognition; Prediction of Stellar Parameters.

 

Resumen

En este artículo se presenta un análisis comparativo de la selección de características usando testores típicos aplicada a la estimación de parámetros estelares. La comparación está basada en la eficiencia de clasificación al usar testores típicos como método de selección de variables contra la eficiencia de clasificación al usar el método de Ramírez (2001), el cual usa algoritmos genéticos. Como clasificador se usó la regla de K–vecinos más cercanos (KNN). Se propuso una modificación al concepto de testor típico para poder aplicarlo en problemas de predicción de una función con imagen en los números reales. En los experimentos, la selección de variables con testores típicos redujo la cantidad de características aproximadamente en un 50%. El error de clasificación usando nuestro método fue menor que usando todas las características o el método de Ramírez.

Palabras Clave: Selección de características, Testores típicos, Reconocimiento lógico combinatorio de patrones, Predicción de parámetros estelares.

 

DESCARGAR ARTÍCULO EN FORMATO PDF

 

Acknowledgement

This work was financially supported by CONACyT (Mexico) through projects J38707–A and I38436–A.

 

References

1. Alba Cabrera E., "New extensions of testor concept for different types of similarity functions" Ph. D. Thesis, ICIMAF, Cuba, 1997. (In Spanish).        [ Links ]

2. Dmitriev A. N., Zhuravlev Y.I. and Krendeliev F.P., "About mathematical principles of objects and phenomena classification", Diskretni Analiz 7 (1966), (In Russian).        [ Links ]

3. Jones L. A., "Star populations in galaxies", Ph. D. Thesis, University of North Carolina, Chapel Hill, North Carolina, 1996.        [ Links ]

4. Jones L.A. and Worthey G., "New age indicators for old stellar populations", Astrophys. J. Lett., 446(1), (1995) L31–L34.        [ Links ]

5. Lazo–Cortes M., Ruiz–Shulcloper J. and Alba–Cabrera E., "An overview of the evolution of the concept of testor", Pattern Recognition, 34(4), (2001), 753–762.        [ Links ]

6. Martínez Trinidad J. F., and Guzmán–Arenas A., "The logical combinatorial approach to pattern recognition an overview through selected works", Pattern Recognition 34(4), (2001), 741–751.        [ Links ]

7. Mitchell T. M., Machine Learning, Carnegie Mellon University, USA, 1976.        [ Links ]

8. Ramirez J. F., Fuentes O., and Gulati R.K., "Prediction of stellar atmospheric parameters using instance–based learning and genetic algorithms", Experimental Astronomy, 12 (3) 163–178, 2001.        [ Links ]

9. Rose J. A., "The integrated spectra of M32 and of 47 Tuc: a comparative study at high spectral resolution", Astronomy J., 107(1), (1994) 206–229.        [ Links ]

10. Ruiz Shulcloper J., Guzmán Arenas A., and Martínez Trinidad J. F., Logical Combinatorial Approach to Pattern Recognition. Feature Selection and Supervised Classification, IPN, Mexico, 1999. (In Spanish).        [ Links ]

11. Worthey G., Faber S. M., Gonzalez J. J., and Burstein D., "Old stellar populations. V. Absorption feature indices for the complete Lick/IDS sample of stars", Astrophys. J., Suppl. Ser 94(2), (1994) 687–722.        [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License