Advertisement

Improving the Performance of Heuristic Algorithms Based on Exploratory Data Analysis

  • Marcela Quiroz C.
  • Laura Cruz-Reyes
  • José Torres-Jiménez
  • Claudia G. Gómez S.
  • Héctor J. Fraire H.
  • Patricia Melin
Part of the Studies in Computational Intelligence book series (SCI, volume 451)

Abstract

This paper promotes the application of empirical techniques of analysis within computer science in order to construct models that explain the performance of heuristic algorithms for NP-hard problems. We show the application of an experimental approach that combines exploratory data analysis and causal inference with the goal of explaining the algorithmic optimization process. The knowledge gained about problem structure, the heuristic algorithm behavior and the relations among the characteristics that define them, can be used to: a) classify instances of the problem by degree of difficulty, b) explain the performance of the algorithm for different instances c) predict the performance of the algorithm for a new instance, and d) develop new strategies of solution. As a case study we present an analysis of a state of the art genetic algorithm for the Bin Packing Problem (BPP), explaining its behavior and correcting its effectiveness of 84.89% to 95.44%.

Keywords

Final Performance Heuristic Algorithm Exploratory Data Analysis Causal Graph Algorithm Behavior 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Garey, M.R., Jonson, D.S.: Computers and Intractability: A Guide to the Theory of NP-Completeness. W. H. Freeman and Company. A classic introduction to the field (1979)Google Scholar
  2. 2.
    Basse, S.: Computer Algorithms, Introduction to Design and Analysis. Editorial Addison-Wesley Publishing Company (1998)Google Scholar
  3. 3.
    Cruz Reyes, L., Nieto-Yáñez, D.M., Rangel-Valdez, N., Herrera Ortiz, J.A., González B, J., Castilla Valdez, G., Delgado-Orta, J.F.: DiPro: An Algorithm for the Packing in Product Transportation Problems with Multiple Loading and Routing Variants. In: Gelbukh, A., Kuri Morales, Á.F. (eds.) MICAI 2007. LNCS (LNAI), vol. 4827, pp. 1078–1088. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  4. 4.
    McGeoch, C.C.: Experimental Analysis of Algorithms. In: Pardalos, P.M., Romeijn (eds.) Handbook of Global Optimization, vol. 2, pp. 489–513 (2002)Google Scholar
  5. 5.
    Hoos, H.H., Stützle, T.: Empirical Analysis of Randomized Algorithms. In: Handbook of Approximation Algorithms and Metaheuristics. Chapman & Hall/CRC, Taylor & Francis Group (2007)Google Scholar
  6. 6.
    Hooker, J.N.: Needed: An empirical science of algorithms. Operations Research 42(2), 201–212 (1994)zbMATHCrossRefGoogle Scholar
  7. 7.
    Barr, S., Golden, L., Kelly, P., Resendez, M., Stewart, R.: Designing and Reporting on Computational Experiments with Heuristic Methods. Journal of Heuristics 1, 9–32 (1995)zbMATHCrossRefGoogle Scholar
  8. 8.
    Cohen, P.R.: Empirical Methods for Artificial Intelligence. The MIT Press, Cambridge (1995)zbMATHGoogle Scholar
  9. 9.
    Lemeire, J.: Learning Causal Models of Multivariate Systems and the Value of it for the Performance Modeling of Computer Programs. PhD thesis. Vrije Universiteit Brussel (2007)Google Scholar
  10. 10.
    Tukey, J.W.: Exploratory Data Analysis. Addison-Wesley (1977)Google Scholar
  11. 11.
    Hartwig, F., Dearing, B.E.: Exploratory Data Analysis. Sage University Paper Series on Quantitative Research Methods, vol. 16. Sage, Newbury Park (1979)Google Scholar
  12. 12.
    Liu, X.: Intelligent Data Analysis: Issues and Challenges. The Knowledge Engineering Review 11, 365–371 (1996)CrossRefGoogle Scholar
  13. 13.
    Quiroz, M.: Caracterización de Factores de Desempeño de Algoritmos de Solución de BPP. Tesis de maestría, Instituto Tecnológico de Cd. Madero, Tamaulipas, México (2009)Google Scholar
  14. 14.
    Beasley, J.E.: OR-library: Distributing test problems by electronic mail. Journal of the Operational Research Society 41(11), 1069–1072 (1990), http://people.brunel.ac.uk/~mastjjb/jeb/orlib/binpackinfo.html Google Scholar
  15. 15.
    Klein, R., Scholl, A.: Bin Packing benchmark data sets, http://www.wiwi.uni-jena.de/Entscheidung/binpp/
  16. 16.
    Euro Especial Interest Group on Cutting and Packing. One Dimensional Cutting and Packing Data Sets, http://paginas.fe.up.pt/~esicup/tiki-list_file_gallery.php?galleryId=1
  17. 17.
    Cutting and Packing at Dresden University. Benchmark data sets, http://www.math.tu-dresden.de/~capad/cpd-ti.html#pmp
  18. 18.
    Pérez, J., Pazos, R.R.A., Frausto, J., Rodríguez, G., Romero, D., Cruz, L.: A Statistical Approach for Algorithm Selection. In: Ribeiro, C.C., Martins, S.L. (eds.) WEA 2004. LNCS, vol. 3059, pp. 417–431. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  19. 19.
    Álvarez, V.: Modelo para representar la Complejidad del problema y el desempeño de algoritmos. Tesis de maestría, Instituto Tecnológico de Cd. Madero, Tamaulipas, México (2006)Google Scholar
  20. 20.
    The TETRAD Project: Causal Models and Statistical Data. TETRAD Homepage, http://www.phil.cmu.edu/projects/tetrad/
  21. 21.
    Alvim, A.C.F., Ribeiro, C.C., Glover, F., Aloise, D.J.: A hybrid improvement heuristic for the onedimensional bin packing problem. Journal of Heuristics 10(2), 205–229 (2004)CrossRefGoogle Scholar
  22. 22.
    Fleszar, K., Charalambous, C.: Average-weight-controlled bin-oriented heuristics for the onedimensional bin-packing problem. European Journal of Operational Research 210(2), 176–184 (2011)MathSciNetzbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Marcela Quiroz C.
    • 1
  • Laura Cruz-Reyes
    • 1
  • José Torres-Jiménez
    • 2
  • Claudia G. Gómez S.
    • 1
  • Héctor J. Fraire H.
    • 1
  • Patricia Melin
    • 3
  1. 1.Instituto Tecnológico de Ciudad MaderoCiudad MaderoMéxico
  2. 2.CINVESTAV-TAMAULIPASVictoriaMéxico
  3. 3.Tijuana Institute of TechnologyTijuanaMéxico

Personalised recommendations