Algorithm of designing compound recognition system on the basis of combining classifiers with simultaneous splitting feature space into competence areas

Abstract

The paper presents the novel adaptive splitting and selection algorithm (AdaSS) used for learning compound pattern recognition system. Splitting a feature space into its constituents and selection of the best area classifier from the pool of available recognizers for each region are key processes of the proposed model. Both take place simultaneously as part of a compound optimization process aimed at maximizing system performance. Evolutionary algorithms are used to find out the optimal solution. The results of experiments for algorithm evaluation purposes prove the quality of the proposed approach.

This is a preview of subscription content, access via your institution.

Fig. 1

References

  1. 1.

    Jain AK, Duin PW, Mao J (2000) Statistical pattern recognition: a review. IEEE Trans PAMI 22(1):4–37

    Google Scholar 

  2. 2.

    Chow CK (1965) Statistical independence and threshold functions. IEEE Trans Electron Comput EC-16:66–68

    Article  Google Scholar 

  3. 3.

    Krzanowski W, Partrige D (1996) Software diversity: practical statistics for its measurement and exploitation. Department of Computer Science, Raport University of Exeter

  4. 4.

    Kuncheva LI, Whitaker CJ (2001) Ten measures of diversity in classifier ensembles: limits for two classifiers. In: IEEE workshop on intelligent sensor processing, Birmingham, pp. 10/1–10/6

  5. 5.

    Xu L, Krzyżak A, Suen Ch Y (1992) Methods of combining multiple classifiers and their applications to handwritting recognition. IEEE Trans Syst Man Cybern 22(3):418–435

    Article  Google Scholar 

  6. 6.

    Kuncheva LI (2004) Combining pattern classifiers: methods and algorithms. Wiley-Interscience, New Jersey

    Google Scholar 

  7. 7.

    Ruta D, Gabrys B (2005) Classifier selection for majority voting. Inf Fusion 6:63–81

    Article  Google Scholar 

  8. 8.

    Kuncheva LI, Whitaker CJ, Shipp CA, Duin RPW (2003) Limits on the majority vote accuracy in classifier fusion. Pattern Anal Appl 6:22–31

    MATH  Article  MathSciNet  Google Scholar 

  9. 9.

    Hashem S (1997) Optimal linear combinations of neural networks. Neural Netw 10(4):599–614

    Article  Google Scholar 

  10. 10.

    Kuncheva LI, Jain LC (2000) Designing classifier fusion systems by genetic algorithms. IEEE Trans Evol Comput 4:327–336

    Article  Google Scholar 

  11. 11.

    Inoue H, Narihisa H (2002) Optimizing a multiple classifier systems. Lect Notes Comput Sci 2417:285–294

    Article  Google Scholar 

  12. 12.

    Wozniak M (2007) Experiments with trained and untrained fusers. In: Corchado E, Corchado JM, Abraham A (eds) Innovations in hybrid intelligent systems. Springer, Berlin, pp 144–150

  13. 13.

    Rao NSV (2004) A generic sensor fusion problem: classification and function estimation. LNCS 3077:16–30

    Google Scholar 

  14. 14.

    Fumera G, Roli F (2005) A theoretical and experimental analysis of linear combiners for multiple classifier systems. IEEE Trans Pattern Anal Mach Intell 27(6):942–956

    Article  Google Scholar 

  15. 15.

    Biggio B et al (2007) Bayesian analysis of linear combiners. LNCS 4472:292–301

    Google Scholar 

  16. 16.

    Tumer K, Ghosh J (1999) Linear and order statistics combiners for pattern classification. In: Sharkley AJC (ed) Combining artificial neural networks. Springer, Berlin, pp 127–155

    Google Scholar 

  17. 17.

    Duin RPW (2002) The combining classifier: to train or not to train? In: Proceedings of the ICPR2002, Quebec City

  18. 18.

    Rastrigin LA, Erenstein RH (1981) Method of collective recognition. Energoizdat, Moscow

    Google Scholar 

  19. 19.

    Baram Y (1998) Partial classification: the benefit of deferrer decision. IEEE Trans Pattern Anal Mach Intell 20(8):769–776

    Article  MathSciNet  Google Scholar 

  20. 20.

    Cordella LP, Foggia P, Sansone C, Tortorella F, Vento M (2000) A cascade multiple expert system for verification. Lect Notes Comput Sci 1857:330–339

    Article  Google Scholar 

  21. 21.

    Giacinto G, Roli F, Fumera G (2000) Design of effective multiple classifier systems by clustering of classifiers. In: Proceedings of 15th international conference on pattern recognition—ICPR’00, vol 2, Barcelona, Spain, pp 160–163

  22. 22.

    Goebel K, Yan W (2004) Choosing classifier for decision fusion. In: Proceedings of the 7th international conference on information fusion, Stockholm, Sweden, pp 562–568

  23. 23.

    Kuncheva LI (2000) Cluster-and-selection method for classifier combination. In: Proceedings of the 4th international conference on knowledge-based intelligent engineering systems and allied technologies, Brighton, pp 185–188

  24. 24.

    Jain AK, Murty MN, Flynn PJ (1999) Data clustering: a review. ACM Comput Surv 31(3):264–323

    Article  Google Scholar 

  25. 25.

    Duda RO, Hart PE, Stork DG (2001) Pattern classification. Wiley-Interscience, New Jersey

    Google Scholar 

  26. 26.

    Devijver PA, Kittler J (1982) Pattern recognition: a statistical approach. Prentice Hall, London

    Google Scholar 

  27. 27.

    Higleyman WH (1962) The design and analysis of pattern recognition experiments. Bell Syst Tech J 41:723–744

    Google Scholar 

  28. 28.

    Kuratowski K, Mostowski A (1996) Set theory. North-Holland Pub. Co, Amsterdam

    Google Scholar 

  29. 29.

    Rokach L (2006) Decomposition methodology for classification tasks—a meta decomposer framework. Pattern Anal Appl 9(2–3):257–271

    MathSciNet  Google Scholar 

  30. 30.

    Gabrys B, Ruta D (2006) Genetic algorithms in classifier fusion. Appl Soft Comput J 6(4):337–347

    Article  Google Scholar 

  31. 31.

    Zhou ZH, Wu J, Tang W (2002) Ensembling neural networks: many could be better than all. Artif Intell 137:239–263

    MATH  Article  MathSciNet  Google Scholar 

  32. 32.

    Duin RPW, Juszczak P, Paclik P, Pekalska E, de Ridder D, Tax DMJ (2004) PRTools4, a matlab toolbox for pattern recognition. Delft University of Technology

  33. 33.

    Siegler RS (1976) Three aspects of cognitive development. Cognit Psychol 8:481–520

    Google Scholar 

  34. 34.

    Newman DJ, Hettich S, Blake CL, Merz CJ (1998) {UCI} Repository of machine learning databases. Dept. of Information and Computer Sciences, University of California, Irvine. http://www.ics.uci.edu/$\sim$mlearn/MLRepository.html

  35. 35.

    Duch W, Grudziński K (2001) Ensembles of similarity-based model. In: Advances in soft computing. Springer, Heidelberg, pp 75–85

  36. 36.

    Duin RPW, Tax DMJ, Experiments with Classifier Combining Rules, Lecture Notes in Computer Science, No. 1857, 2000, pp 16–29

  37. 37.

    Kuncheva LI, Bezdek JC, Duin RPW (2001) Decision templates for multiple classifier fusion: an experimental comparision. Pattern Recognition 34:299–314

    MATH  Article  Google Scholar 

  38. 38.

    Van Erp M, Vuurpijl LG, Schomaker LRB, An overview and comparison of voting methods for pattern recognition. In: Proc. of the 8th International Workshop on Frontiers in Handwriting Recognition (IWFHR.8), Niagara-on-the-Lake, Canada, pp 195–200

Download references

Acknowledgments

This work is supported by the grant of The Polish State Committee for Scientific Research (2006–2009).

Author information

Affiliations

Authors

Corresponding author

Correspondence to Michal Wozniak.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Jackowski, K., Wozniak, M. Algorithm of designing compound recognition system on the basis of combining classifiers with simultaneous splitting feature space into competence areas. Pattern Anal Applic 12, 415 (2009). https://doi.org/10.1007/s10044-008-0137-7

Download citation

Keywords

  • Pattern recognition
  • Multiple classifier system
  • Clustering
  • Selection
  • Evolutionary algorithms