Skip to main content
SpringerLink
Log in

A New Fuzzy Support Vector Machine Based on the Weighted Margin

  • Published:
Neural Processing Letters Aims and scope Submit manuscript

Abstract

The ideas from fuzzy neural networks and support vector machine (SVM) are incorporated to make SVM classifiers perform better. The influence of the samples with high uncertainty can be decreased by employing the fuzzy membership to weigh the margin of each training vector. The linear separability, fuzzy margin, optimal hyperplane, generalization and soft fuzzy margin algorithms are discussed. A new optimization problem is obtained and SVM is then completely reformulated into a new fuzzy support vector machine (NFSVM). Moreover, the generation bound of NFSVM can be described. We also introduce the membership function in fuzzy neural networks to do some experiments. The results demonstrate that the proposed NFSVM can produce better results than regular SVM and Fuzzy Kernel Perceptron (FKP) in some real cases.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Burges, C. J. C.: A tutorial on support vector machines for pattern recognition. Data Mining and Knowledge Discovery, 2 (1998) 121–167.

    Google Scholar 

  2. Chen, J. and Chen, C. Fuzzy kernel perceptron, IEEE Trans Neural Networks, 13(6) (2002)1364–1373.

    Google Scholar 

  3. Cristianini, N. and Schawe-Taylor, J.: An Introduction to Support Vector Machines, Cambridge, Cambridge University Press, 2000.

    Google Scholar 

  4. Keller, J. M. and Hunt, D. J.: Incorporating fuzzy membership functions into the perceptron algorithm, IEEE Trans Pattern Anal. Mach. Intelli. 7 (1985) 693–699.

    Google Scholar 

  5. Lin, C. and Wang, S.: Fuzzy Support Vector Machines, IEEE Trans. Neural Networks, 13(2) (2002) 464–471.

    Google Scholar 

  6. Ra¨ tsch, G.: Robust boosting via convex optimization. Ph.D thesis, University of Posdam. 2001.

  7. Schawe-Taylor, J. and Cristianini, N.: On the generalization of soft margin algorithms, IEEE Trans. Inform. Theory, 48(10) (2002) 2721–2735.

    Google Scholar 

  8. Scho¨ lkopf, B. and Smola, A. J.: Learning with kernels MIT Press. 2002. Cambridge, MA

  9. Valiant, L. G. A Theory of the Learnable, Communi. ACM, 27(11) (1984) 1134–1142.

    Google Scholar 

  10. Vapnik, V.: The Nature of Statistical Learning Theory. New York: Springer-verlag, 1995.

    Google Scholar 

  11. Vapnik, V.: Statistical Learning Theory, Addison-Wesley, 1998.

Download references

Author information

Authors and Affiliations

  1. New Star Research Institute of Applied Technology, Hefei, 230031, P. R. China

    Qing Tao

  2. Laboratory of Complex Systems and Intelligence Science, Institute of Automation, Chinese Academy of Sciences, Beijing, 100080, P. R. China

    Jue Wang

Authors
  1. Qing Tao
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jue Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Tao, Q., Wang, J. A New Fuzzy Support Vector Machine Based on the Weighted Margin. Neural Processing Letters 20, 139–150 (2004). https://doi.org/10.1007/s11063-004-1640-5

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11063-004-1640-5

Search

Navigation