ISSN 0869-6632 (Print)
ISSN 2542-1905 (Online)

For citation:

Kovalchuk A. V., Bellyustin N. S. Classification algorithm of streaming signals based on the online support vector machine. Izvestiya VUZ. Applied Nonlinear Dynamics, 2015, vol. 23, iss. 5, pp. 62-79. DOI: 10.18500/0869-6632-2015-23-5-62-79

This is an open access article distributed under the terms of Creative Commons Attribution 4.0 International License (CC-BY 4.0).
Full text PDF(Ru):
(downloads: 64)
Article type: 

Classification algorithm of streaming signals based on the online support vector machine

Kovalchuk Andrej Viktorovich, Institute of Applied Physics of the Russian Academy of Sciences
Bellyustin Nikolaj Sergeevich, Federal state budgetary scientific institution "Scientific-research radiophysical Institute"

The work proposed a modification of support vector machines (SVM) to train and classify in real time (online) streams of data. The algorithm is tested on the data handwriting figures and shown that its error is comparable to SVM direct solution error. Speed and support vectors number of proposed SVM algorithm is smaller than in other known SVM implementations. Finally, a ternary classificator for 2-class problem is proposed which shows better results than binary.

  1. Cristianini N., J. Shawe-Taylor J. An Introduction to Support Vector Machines and Other Kernel-Based Learning Methods. Cambridge University Press, 2000.
  2. Makarenko A.V., Pravdivtsev A.V., Volovik M.G. Problems of modeling and analysis of infrared thermo maps human brain // Izvestiya VUZ. Applied Nonlinear Dynamics. 2011. Vol. 19, No 6. P. 145. (in Russian).
  3. Cauwenberghs G., Poggio T. Incremental and Decremental Support Vector Machine Learning // Advances in Neural Information Processing Systems, 2000. P. 409.
  4. Bordes A., Ertekin S., Weston J., Bottou L. Fast kernel classifiers with online and active learning // Journal of Machine Learning Research. 2005. Vol. 6. P. 1579.
  5. Roobaert D. DirectSVM: A Simple Support Vector Machine Perceptron. VLSI Signal Processing, 2002. P. 147.
  6. Orabona F., Castellini C., Caputo B., Jie L., Sandini G. On-line independent support vector machines // Pattern Recognition. 2010. Vol. 43. P. 1402.
  7. Schoelkopf B., Smola A. New Support Vector Algorithms // NeuroCOLT Technical Report NC2-TR-1998-031. 1998.
  8. Algorithms and programs for dependencies reconstraction / Ed. V.N. Vapnik. M.: Physmatlit, 1984 (in Russian).
  9. Vapnik V. Estimation of Dependences Based on Empirical Data. Berlin: Springer- Verlag, 1982.
  10. Platt J. Fast training of support vector machines using sequential minimal optimization / Eds B. Scholkopf, C.J.C. Burges, A.J. Smola. Advances in Kernel Methods – Support Vector Learning. Cambridge, MA: MIT Press, 1999.
  11. Rastrigin L.A., Erenstein R.H. The method of collective recognition. Energoizdat, 1981. 80 p. (in Russian).
  12. Chang C.-C., Lin C.-J. LIBSVM: A Library for Support Vector Machines. Technical report // Computer Science and Information Engineering. National Taiwan University, 2001–2004.
  13. Collobert R., Bengio S. SVMTorch: Support vector machines for large-scale regression problems // Journal of Machine Learning Research. 2001. Vol. 1. P. 143.
  14. Keerthi S.S., Gilbert E.G. Convergence of a generalized SMO algorithm for SVM classifier design // Machine Learning. 2002. Vol. 46. P. 351.
Short text (in English):
(downloads: 120)