Inglés (pdf)
Articulo en XML
Referencias del artículo
Traducción automática
Enviar articulo por email
Permalink
Ver contenido completo en pdf.
Servicios Personalizados
Revista
Articulo
Indicadores
-
Citado por SciELO -
Accesos
Links relacionados
-
Similares en
SciELO
Compartir
Revista de Matemática Teoría y Aplicaciones
versión impresa ISSN 1409-2433
Rev. Mat vol.23 no.2 San José jul./dic. 2016
Artículos
Clustering problems in a multiobjective framework
Problemas de agrupación en un ambiente multiobjetivo
1Department of Interdisciplinary Mathematics, Institute of Cybernetics, Mathematics andPhysics, Havana, Cuba. E-Mail: yunay@icimaf.cu
2Department of Interdisciplinary Mathematics, Institute of Cybernetics, Mathematics andPhysics, Havana, Cuba. E-Mail: rbeau3105@gmail.com, rbeausol@icimaf.cu
We propose a new algorithm using tabu search to deal with biobjective clustering problems. A cluster is a collection of records that are similar to one other and dissimilar to records in other clusters. Clustering has applications in VLSI design, protein-protein interaction networks, data mining and many others areas. Clustering problems have been subject of numerous studies; however, most of the work has focused on single-objective problems. In the context of multiobjective optimization our aim is to find a good approximation to the Pareto front and provide a method to make decisions. As an application problem we present the zoning problem by allowing the optimization of two objectives.
Keywords: combinatorial data analysis; clustering; tabu search; multiobjective optimization
En este trabajo proponemos un nuevo algoritmo usando un enfoque de búsqueda tabú para dar solución a problemas de agrupación (clusters) tomando en consideración dos objetivos. La tarea de agrupación se refiere a la agrupación de objetos, observaciones, o casos. Una agrupación es una colección de objetos similares entre sí y disímiles entre agrupaciones. Aplicaciones de agrupaciones tienen lugar en los diseños VLSI, redes de interacción proteina-proteina, minería de datos y muchas otras áreas. Los problemas de agrupación han sido ampliamente estudiados, pero su descripción se ha basado en la consideración de solamente un objetivo. En el contexto de optimización multiobjetivo nuestro objetivo es hallar una buena aproximación de la frontera Pareto y proveer un método para la toma de decisión. Como aplicación presentamos el problema de zonificación optimizando dos objetivos.
Palabras clave: Análisis de datos combinatorio; cluster; búsqueda tabú; optimización multiobjetivo
Acknowledgements
The authors acknowledge B. Bernábe-Loranca for providing the data set in the zoning problem.
References
Aeberhard, S.; Coomans, D.; de Vel, O.(1994) "Comparative analysis of statistical pattern recognition methods in high dimensional settings", Pattern Recognition 27(8): 1065-1077. [ Links ]
Beausoleil, R.P.(2001) "Multiple criteria scatter search", in: Proceedings MIC'2001- 4th Metaheuristics International Conference, Porto, Portugal: 539-543. [ Links ]
Bernábe, B.; Coello, C.A.;Osorio, M. (2012) "A multiobjective approach for the heuristic optimization of compactness and homogeneity in the optimal zoning", Journal of Applied Research and Technology10(3): 447-457. [ Links ]
Brusco, M.J.; Cradit, J.D. (2005) "Bicriterion methods for partitioning dissimilarity matrices", British Journal of Mathematical and Statistical Psychology58(2): 319-332. [ Links ]
Brusco, M.J.; Stahl, S. (2005) Branch-and-Bound Applications in Combinatorial Data Analysis. Springer, New York. [ Links ]
Caballero, R.; Laguna, M.; Martí, R.; Molina, J. (2006) "Multiobjective clustering with metaheuristic optimization technology", Technical Report 02-06University of ColoradoCO, available in: http://www.uv.es/sestio/TechRep/ tr02-06.pdf [ Links ]
Delattre, M.; Hansen, P. (1980) "Bicriterion cluster analysis", IEEE Transaction on Pattern Analysis and Machine Intelligence2(4): 227-291. [ Links ]
Fisher, R.A.(1936) "The use of multiple measurements in taxonomic problems", Annals of Eugenics7(2): 179-188. [ Links ]
Glover, F. (1986) "Future paths for integer programming and links to artificial intelligence", Computers and Ops. Res.13(5): 533-549. [ Links ]
Handl, J.; Knowles, J.(2004) "Evolutionary multiobjective clustering", in: X.Yaoet al. (Eds.) Parallel Problem Solving from Nature (PPSN VIII), Lecture Notes on Computer Science 3242, Springer, Heidelberg: 1081-1091. [ Links ]
Hansen, P.; Jaumard, B.(1997) "Cluster analysis and mathematical programming", Mathematical Programming79(1-3): 191-215. [ Links ]
Jain, A.K.; Dubes, R.C.(1988) Algorithms for Clustering Data. PrenticeHall, Upper Saddle River NJ. [ Links ]
Kurgan, L.A.; Cios, K.J.; Tadeusiewicz, R.; Ogiela, M.; Goodenday, L.S. (2001) "Knowledge discovery approach to automated cardiac SPECT diagnosis", Artificial Intelligence in Medicine23(2): 149-169. [ Links ]
Mc Queen, J. (1967) "Some methods for classification and analysis of multivariate observations", in: Proceedings of the Fifth Berkeley Symposiums on Mathematical Statistics and Probability1: 281-297. [ Links ]
Pacheco, J.; Valencia, O.(2003) "Design of hybrids for the minimum sumof-squares clustering problem", Computational Statistics and Data Analysis 43(2): 235-248. [ Links ]
Rand, W.M.(1971) "Objective criteria for the evaluation of clustering methods", J. Am. Stat. Assoc66(336): 846-850. [ Links ]
Reeves, C.R. (Ed.) (1993) Modern Heuristics Techniques for Combinatorial Problems. Wiley, New York. [ Links ]
Vendramin, L.; Campello, R.J.G.B.; Hruschka, E.R. (2010) "Relative clustering validity criteria: A comparative overview", Statistical Analysis and Data Mining 3(4): 209-235. [ Links ]
Received: October 14, 2014; Revised: September 09, 2015; Accepted: February 20, 2016
This is an open-access article distributed under the terms of the Creative Commons Attribution License
