Combining Dissimilarities for Three–Way Data Classification
Combinación de disimilitudes para la clasificación de datos de tres vías
Diana Porro Muñoz1,2, Isneri Talavera1, Robert P. W. Duin2, and Mauricio Orozco Alzate3
1 Advanced Technologies Application Center (CENATAV), Cuba.
2 Pattern Recognition Lab., TU Delft, The Netherlands.
]]> 3 Universidad Nacional de Colombia Sede Manizales, Colombia. E–mail: dporro@cenatav.co.cu, italavera@cenatav.co.cu, r.duin@ieee.org, morozcoa@bt.unal.edu.co
Article received on February 28, 2011.
Accepted on June 30, 2011.
Abstract
The representation of objects by multidimensional arrays is widely applied in many research areas. Nevertheless, there is a lack of tools to classify data with this structure. In this paper, an approach for classifying objects represented by matrices is introduced, based on the advantages and success of the combination strategy, and particularly in the dissimilarity representation. A procedure for obtaining the new representation of the data has also been developed, aimed at obtaining a more powerful representation. The proposed approach is evaluated on two three–way data sets. This has been done by comparing the different ways of achieving the new representation, and the traditional vector representation of the objects.
Keywords: Classification, three–way data, combination and dissimilarity representation.
Resumen
]]> La representación de objetos a través de arreglos multidimensionales es ampliamente utilizada en muchas áreas de investigación. Sin embargo, el desarrollo de herramientas para clasificar datos con dicho tipo de estructura ha sido insuficiente. En este trabajo se introduce una metodología para clasificar objetos que son representados por matrices, basada en las ventajas y éxitos de la estrategia de combinación y particularmente en la representación por disimilitudes. También se propone el procedimiento para obtener la nueva representación de los datos. La propuesta realizada en este trabajo se evaluó en dos conjuntos de datos tres–vías. Esta evaluación se realizó mediante la comparación entre las diferentes maneras de obtener la nueva representación, y la representación tradicional de los objetos a través de vectores.Palabras clave: Clasificación, datos de tres–vías, combinación y representación por disimilitudes.
DESCARGAR ARTÍCULO EN FORMATO PDF
Acknowledgment
We acknowledge financial support from the FET programme within the EU FP7, under the project "Similarity–based Pattern Analysis and Recognition– SIMBAD" (contract 213250). We would also like to thank to the project "Cálculo científico para caracterización e identificación en problemas dinámicos" (code Hermes 10722) granted by Universidad Nacional de Colombia.
References
1. Ballabio, D., Consonni, V. & Todeschini, R. (2007). Classification of multiway analytical data based on MOLMAP approach. Analytica Chimica Acta, 605 (2), 134–146. [ Links ]
2. Benbrahim, M., Daoudi, A., Benjelloun, K. & Ibenbrahim, A. (2005). Discrimination of Seismic Signals Using Artificial Neural Networks. World Academy of Science, Engineering and Technology, Istanbul, Turkey, (4), 4–7. [ Links ]
3. Damoulas, T. & Girolami, M. A. (2009). Combining feature spaces for classification. Pattern Recognition, 42 (11), 2671–2683. [ Links ]
4. Ibba, A., Duin, R. P. W. & Lee, W. J. (2010). A study on combining sets of differently measured dissimilarities. 20th International conference on Pattern Recognition, Istambul, Turkey, 3360–3363. [ Links ]
5. Kim, S. W. & Duin, R. P. W. (2007). On Combining Dissimilarity–Based Classifiers to Solve the Small Sample Size Problem for Appearance–Based Face Recognition. Advances in Artificial Intelligence. Lecture Notes in Computer Science, 4509, 110–121. [ Links ]
6. Kittler, J., Hatef, M., Duin, R. P. W. & Matas, J. (1998). On combining classifiers. IEEE Transactions on Pattern Analysis and Machine Intelligence, 20 (3), 226–239. [ Links ]
7. Kroonenberg, P. M. (2008). Applied Multiway Data Analysis. Hoboken, New Jersey; Wiley–Interscience [ Links ]
8. Kuncheva, L. I. (2004). Combining Pattern Classifiers: Methods and Algorithms. Hoboken, New Jersey: John Wiley [ Links ]
9. Orozco–Alzate, M., García, M. E., Duin, R. P. W. & Castellanos, C. G. (2006). Dissimilarity–Based classification of seismic signals at Nevado del Ruiz Volcano. Earth Sciences Research Journal, 10 (2), 57–65. [ Links ]
10. Paclik, P. & Duin, R. P. W. (2003). Dissimilarity–based classification of spectra: computational issues. Real Time Imaging, 9 (4), 237–244. [ Links ]
11. Pekalska, E. & Duin, R. P. W. (2001). On Combining Dissimilarity Representations. Second International Workshop on Multiple Classifier Systems. Lecture Notes in Computer Science, 2096, 359–368. [ Links ]
12. Pekalska, E. & Duin, R. P. W. (2005). The Dissimilarity Representation for Pattern Recognition. Foundations and Applications. Hackensack, N.J.: World Scientific. [ Links ]
13. Porro–Muñoz, D., Talavera, I. & Duin, R. P. W. (2009). Multi–way data analysis (RNPS No.2142 Reporte técnico, serie Azul RT_014). Cuba: CENATAV. [ Links ]
14. Porro–Muñoz, D., Duin, R. P. W., Talavera, I. & Hdez., N. (2009). The Representation of Chemical Spectral Data for Classification. Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. Lecture Notes in Computer Science, 5856, 513–520. [ Links ]
15. Porro–Muñoz, D., Duin, R. P. W., Orozco–Alzate, M, Talavera–Bustamante, I. & Londono–Bonilla, J.M (2010). The Dissimilarity Representation as a Tool for Three–Way Data Classification: A 2D Measure. Structural, Syntactic and Statistical Pattern Recognition. Lecture Notes in Computer Science, 6218, 569–578. [ Links ]
16. Sadecka, J. & Tothova, J. (2007). Fluorescence spectroscopy and chemometrics in the food classification – a review. Czech Journal of Food Sciences, 25 (4), 159–173. [ Links ]
17. Skov, T., Ballabio, D. & Bro, R. (2008). Multiblock Variance Partitioning. A new approach for comparing variation in multiple data blocks. Analytica Chimica Acta, 615 (1), 18–29. Retrieved from: http://models.kvl.dk/Wine_GCMS_FTIR. [ Links ]
]]>