Journal of the Operational Research Society

, Volume 60, Issue 8, pp 1116–1122 | Cite as

Support vector machine learning with an evolutionary engine

  • R Stoean
  • M Preuss
  • C Stoean
  • E El-Darzi
  • D Dumitrescu
Special Issue Paper

Abstract

The paper presents a novel evolutionary technique constructed as an alternative of the standard support vector machines architecture. The approach adopts the learning strategy of the latter but aims to simplify and generalize its training, by offering a transparent substitute to the initial black-box. Contrary to the canonical technique, the evolutionary approach can at all times explicitly acquire the coefficients of the decision function, without any further constraints. Moreover, in order to converge, the evolutionary method does not require the positive (semi-)definition properties for kernels within nonlinear learning. Several potential structures, enhancements and additions are proposed, tested and confirmed using available benchmarking test problems. Computational results show the validity of the new approach in terms of runtime, prediction accuracy and flexibility.

Keywords

evolutionary algorithms support vector machines classification regression 

References

  1. Beielstein T (2006). Experimental Research in Evolutionary Computation—The New Experimentalism. Natural Computing. Springer-Verlag: Berlin.Google Scholar
  2. Bosch RA and Smith JA (1998). Separating hyperplanes and the authorship of the disputed federalist papers. Amer Math Month 105(7): 601–608.CrossRefGoogle Scholar
  3. Cortes C and Vapnik V (1995). Support vector networks. J Mach Learn 20: 273–297.Google Scholar
  4. Eiben AE and Smith JE (2003). Introduction to Evolutionary Computing. Springer-Verlag: Berlin.CrossRefGoogle Scholar
  5. Friedrichs F and Igel C (2004). Evolutionary tuning of multiple svm parameters. In: Verleysen M. (ed). Proceedings of the 12th ESANN, Bruges, Belgium. D-Side Publications: Evere, Belgium, pp. 519–524.Google Scholar
  6. Howley T and Madden MG (2005). The genetic kernel support vector machine: Description and evaluation. Artif Intell Rev 24(3–4): 379–395.CrossRefGoogle Scholar
  7. Hsu CW and Lin CJ (2002). A comparison of methods for multi-class support vector machines. IEEE Trans NN 13(2): 415–425.CrossRefGoogle Scholar
  8. Jun SH and Oh KW (2006). An evolutionary statistical learning theory. Comput Intell 3(3): 249–256.Google Scholar
  9. Mierswa I (2006). Making indefinite kernel learning practical. Technical Report, University of Dortmund.Google Scholar
  10. Smola AJ and Scholkopf B (1998). A tutorial on support vector regression. Technical Report, University of London.Google Scholar
  11. Stoean R, Preuss M, Dumitrescu D and Stoean C (2006). Evolutionary support vector regression machines. In: O'Conner L. (ed). IEEE Postproceedings of SYNASC, Timisoara, Romania. IEEE Press: Los Alamitos, CA, pp. 330–335.Google Scholar
  12. Stoean R, Preuss M, Stoean C and Dumitrescu D (2007). Concerning the potential of evolutionary support vector machines. In: Srinivasan D. and Wang L. (eds). Proceedings of the IEEE CEC, Singapore. IEEE Press: Piscataway, NJ, pp. 1436–1443.Google Scholar
  13. Vapnik V (1998). Statistical Learning Theory. Wiley: New York.Google Scholar

Copyright information

© Palgrave Macmillan 2008

Authors and Affiliations

  • R Stoean
    • 1
  • M Preuss
    • 2
  • C Stoean
    • 1
  • E El-Darzi
    • 3
  • D Dumitrescu
    • 4
  1. 1.University of CraiovaCraiovaRomania
  2. 2.University of DortmundDortmundGermany
  3. 3.University of WestminsterLondonUK
  4. 4.University of Cluj-NapocaClujRomania

Personalised recommendations