VARIABLE SUBSET SELECTION FOR BRAIN-COMPUTER INTERFACE - PCA-based Dimensionality Reduction and Feature Selection

N. S. Dias, M. Kamrunnahar, P. M. Mendes, S. J. Schiff, J. H. Correia

2009

Abstract

A new formulation of principal component analysis (PCA) that considers group structure in the data is proposed as a Variable Subset Selection (VSS) method. Optimization of electrode channels is a key problem in brain-computer interfaces (BCI). BCI experiments generate large feature spaces compared to the sample size due to time limitations in EEG sessions. It is essential to understand the importance of the features in terms of physical electrode channels in order to design a high performance yet realistic BCI. The VSS produces a ranked list of original variables (electrode channels or features), according to their ability to discriminate between tasks. A linear discrimination analysis (LDA) classifier is applied to the selected variable subset. Evaluation of the VSS method using synthetic datasets selected more than 83% of relevant variables. Classification of imagery tasks using real BCI datasets resulted in less than 16% classification error.

References

  1. Babiloni, C., Carducci, F., Cincotti, F., Rossini, P.M., Neuper, C., Pfurtscheller G., Babiloni, F., 1999. Human movement-related potentials vs desynchronization of EEG alpha rhythm: A highresolution EEG study. In NeuroImage, vol. 10, pp. 658-665.
  2. Dias, N.S., Kamrunnahar, M., Mendes, P.M., Schiff, S.J., Correia, J.H. Customized Linear Discriminant Analysis for Brain-Computer Interfaces. In Proc. CNE 7807 IEEE/EMBS 2-5 May 2007, pp. 430-433.
  3. Dias, N.S., Kamrunnahar, M., Mendes, P.M., Schiff, S.J., Correia, J.H. Comparison of EEG Pattern Classification Methods for Brain-Computer Interfaces. In Proc.29th EMBC 22-26 Aug 2007, pp.2540-2543.
  4. Dillon, W.R., Mulani, N., Frederick, D.G., 1989. On the Use of Component Scores in the Presence of Group Structure. JOURNAL OF CONSUMER RESEARCH, vol. 16, pp. 106-112.
  5. Duda, R.O., Hart, P.E., Stork, D.G., 2000. Pattern Classification. Wiley.
  6. John, G.H., Kohavi, R., Pfleger, K., 1994. Irrelevant Features and the Subset Selection Problem. In Proc. 11th Int. Conf. Machine Learning, 121-129.
  7. Jolliffe, I.T., 2002. Principal Component Analysis, Springer 2nd edition.
  8. Kamrunnahar, M., Dias, N.S., Schiff, S.J. Model-based Responses and Features in Brain Computer Interfaces. In Proc. 30th IEEE EMBC 20-25 Aug 2008, pp.4482- 4485.
  9. Lai, C., Reinders, M.J.T., Wessels, L., 2006. Random subspace method for multivariate feature selection. Pattern Recognition Letters, no.27, pp.1067-1076.
  10. Schiff, S.J., Sauer, T., Kumar, R., Weinstein, S.L., 2005. Neuronal spatiotemporal pattern discrimination: The dynamical evolution of seizures. NEUROIMAGE, 28 ed, pp. 1043-1055.
  11. Wolpaw, J.R., McFarland, D.J., Vaughan, T.M., 2000. Brain-Computer Interface Research at the Wadsworth Center. IEEE TRANSACTIONS ON REHAB. ENGINEERING, vol. 8, no. 2, pp. 222-226.
  12. Yu, L., Liu, H., 2004. Efficient feature Selection via Analysis of Relevance and Redundancy. Journal of Machine Learning, no. 5, pp. 1205-1224.
  13. 11. k = number of first %rAGV elements whose cumulative sum > d
  14. 12. for j=1 to p do
  15. 13. TruncVarj = Si=1,..,k r?i×rVji2
  16. 14. end for
  17. 15. [rTruncVar,Index] = sort TruncVar in descending order
  18. 16. sub = 1st k elements in Index
Download


Paper Citation


in Harvard Style

Dias N., Kamrunnahar M., Mendes P., Schiff S. and Correia J. (2009). VARIABLE SUBSET SELECTION FOR BRAIN-COMPUTER INTERFACE - PCA-based Dimensionality Reduction and Feature Selection . In Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2009) ISBN 978-989-8111-65-4, pages 35-40. DOI: 10.5220/0001533200350040


in Bibtex Style

@conference{biosignals09,
author={N. S. Dias and M. Kamrunnahar and P. M. Mendes and S. J. Schiff and J. H. Correia},
title={VARIABLE SUBSET SELECTION FOR BRAIN-COMPUTER INTERFACE - PCA-based Dimensionality Reduction and Feature Selection},
booktitle={Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2009)},
year={2009},
pages={35-40},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001533200350040},
isbn={978-989-8111-65-4},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2009)
TI - VARIABLE SUBSET SELECTION FOR BRAIN-COMPUTER INTERFACE - PCA-based Dimensionality Reduction and Feature Selection
SN - 978-989-8111-65-4
AU - Dias N.
AU - Kamrunnahar M.
AU - Mendes P.
AU - Schiff S.
AU - Correia J.
PY - 2009
SP - 35
EP - 40
DO - 10.5220/0001533200350040