Using Probabilistic Direct Multi-class Support Vector Machines to Improve Mental States Based-Brain Computer Interface

  • Mounia HendelEmail author
  • Fatiha Hendel
Conference paper
Part of the Smart Innovation, Systems and Technologies book series (SIST, volume 111)


Brain-Computer Interface (BCI) system allows physically challenged people to operate with their external surroundings, just through their brain signals. Since the objective of BCI is to categorize the brain signals into homogeneous classes each of which represents a mental state, it is necessary to choose an appropriate discrimination approach. So, we use the Support Vector Machines (SVM) due to their multiple benefits. The SVM are suggested to treat binary problems, their conversion to multiclass cases (M-SVM) includes: indirect methods based on decomposition approaches, and direct methods that consider all classes simultaneously. This experiment aims to introduce the use of the four existing direct M-SVM in the problematic of mental states recognition. The discriminators operate independently and give probability estimates relative to five mental states. Results indicate that models generate nearly similar accuracies. Nevertheless, with an average rates ranging from 68.25% to 90.86%, Crammer and Singer discriminator outperforms the other models.


Direct M-SVM BCI EEG DWT Mental tasks 


  1. 1.
    Vaid, R.S., Singh, P., Kaur, C.: EEG signal analysis for BCI interface: a review. In: IEEE Transaction on Advanced Computing and Communication Technologies, pp. 143–147 (2015)Google Scholar
  2. 2.
    Prashant, P., Joshi, A., Gandhi, V.: Brain computer interface: a review. In: 5th Nirma University International Conference on Engineering, pp. 1–6. IEEE (2015)Google Scholar
  3. 3.
    Gupta, A., Agrawal, R.K., Kaur, B.: Performance enhancement of mental task classification using EEG signal: a study of multivariate feature selection methods. Soft Comput. 19, 2799–2812 (2015)CrossRefGoogle Scholar
  4. 4.
    Hendel, M., Benyettou, A., Hendel, F.: Hybrid self organizing map and probabilistic quadratic loss multi-class support vector machine for mental tasks classification. Inform. Med. Unlocked 4, 1–9 (2016)CrossRefGoogle Scholar
  5. 5.
    Gupta, A., Kirar, J.S.: A novel approach for extracting feature from EEG signal for mental task classification. In: IEEE Transaction on Computing and Network Communications, pp. 829–832 (2015)Google Scholar
  6. 6.
    Gupta, A., Kumar, D.: Fuzzy clustering-based feature extraction method for mental task classification. Brain Inform. 4, 135–145 (2016)CrossRefGoogle Scholar
  7. 7.
    El Bahy, M.M., Hosny M., Mohamed, W.A., Ibrahim, M.: EEG signal classification using neural network and support vector machine in brain computer interface. In: Advances in Intelligent Systems and Computing, vol. 533, pp. 246–256. Springer (2017)Google Scholar
  8. 8.
    Liang, N., Saratchandran, P., Huang, G., Sundararajan, N.: Classification of mental tasks from EEG signals using extreme learning machine. Int. J. Neural Syst. 16(1), 29–38 (2006)CrossRefGoogle Scholar
  9. 9.
    Weston, J., Watkins, C.: Multi-class support vector machines. Royal Holloway, University of London, Department of Computer Science, Technical report CSD-TR-98-04 (1998)Google Scholar
  10. 10.
    Crammer, K., Singer, Y.: On the algorithmic implementation of multiclass kernel-based vector machines. J. Mach. Learn. Res. 2, 265–292 (2001)zbMATHGoogle Scholar
  11. 11.
    Lee, Y., Lin, Y., Wahba, G.: Multicategory support vector machines: theory and application to the classification of microarray data and satellite radiance data. J. Am. Stat. Assoc. 99(465), 67–81 (2004)MathSciNetCrossRefGoogle Scholar
  12. 12.
    Guermeur, Y., Monfrini, E.: A quadratic loss multi-class SVM for which a radius-margin bound applies. Informatica 22(1), 73–96 (2011)MathSciNetzbMATHGoogle Scholar
  13. 13.
    Keirn, Z.: Alternative modes of communication between man and machines. Master’s dissertation, Department of Electrical Engineering, Purdue University, USA (1988)Google Scholar
  14. 14.
  15. 15.
    Keirn, Z., Aunon, J.: A new mode of communication between man and his surroundings. IEEE Trans. Biomed. Eng. 37(12), 1209–1214 (1990)CrossRefGoogle Scholar
  16. 16.
    Palaniappan, R.: Utilizing gamma band to improve mental task based brain-computer interface designs. IEEE Trans. Neural Syst. Rehabil. Eng. 14(3), 299–303 (2006)CrossRefGoogle Scholar
  17. 17.
    Diez, P.F., Mut, V., Laciar, E., Torres, A., Avila, E.: Application of the empirical mode decomposition to the extraction of features from EEG signals for mental task classification. In: Engineering in Medicine and Biology Society, Minneapolis, pp. 2579–2582 (2009)Google Scholar
  18. 18.
    Tolic, M., Jovic, F.: Classification of wavelet transformed eeg signals with neural network for imagined mental and motor tasks. 45(1), 130–138 (2013)Google Scholar
  19. 19.
    Hariharan, H., Vijean, V., Sindhu, R., Divakar, P., Saidatul, A., Yaacob, Z.: Classification of mental tasks using stockwell transform. Comput. Electr. Eng. 40, 1741–1749 (2014)CrossRefGoogle Scholar
  20. 20.
    Guermeur, Y.: A generic model of multi-class support vector machine. Int. J. Intell. Inf. Database Syst. 6(6), 555–577 (2012)CrossRefGoogle Scholar
  21. 21.
    Lauer, F., Guermeur, Y.: MSVMpack: a multi-class support vector machine package. J. Mach. Learn. Res. 12, 2269–2272 (2011)Google Scholar
  22. 22.
    Bennani, Y., Bossaert, F.: Predictive neural networks for traffic disturbance detection in the telephone network. In: Proceedings of IMACS-IEEE Computational Engineering in System Applications, France (1996)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Department of Computer ScienceUniversity of Science and Technology (USTO-MB)OranAlgeria
  2. 2.Higher School of Electrical Engineering and Energetic (ESG2E)OranAlgeria
  3. 3.Department of ElectronicUniversity of Science and Technology (USTO-MB)OranAlgeria

Personalised recommendations