Evolutionary Weighted Ensemble for EEG Signal Recognition
Recognition of an EEG signal is a very complex but very important problem. In this paper we focus on a simplified classification problem which consists of detection finger movement based on an analysis of seven EEG sensors. The signals gathered by each sensor are subsequently classified by the respective classification algorithm, which is based on data compression and so called LZ-Complexity. To improve overall accuracy of the system, the Evolutionary Weighted Ensemble (EWE) system is proposed. The parameters of the EWE are set in a learning procedure which uses an evolutionary algorithm tailored for that purpose. To take full advantage of information returned by sensor classifiers, setting negative weights are permitted, which significantly raises overall accuracy. Evaluation of EWE and its comparison against selected traditional ensemble algorithm is carried out using empirical data consisting of almost 5 hundred samples. The results show that the EWE algorithm exploits the knowledge represented by the sensor classifiers very effectively, and greatly improves classification accuracy.
KeywordsEEG classification machine learning ensemble of classifiers evolutionary algorithms
Unable to display preview. Download preview PDF.
- 2.Chow, C.K.: Statistical independence and threshold functions. IEEE Transaction on Electronic Computers, EC-16, 66–68 (1965)Google Scholar
- 5.Freund, Y.: Boosting week learning algorithm by majority. Information and Computation 121(2) (1995)Google Scholar
- 6.Freund, Y., Schapire, R.E.: Experiments with a New Boosting Algorithm. In: Proceedings of the International Conference on Machine Learning, pp. 148–156 (1996)Google Scholar
- 13.Alpaydin, E.: Introduction to Machine Learning. The MIT Press (2004)Google Scholar
- 14.Back, T., Fogel, D., Michalewicz, Z.: Handbook of Evolutionary Computation, Oxford Univ. Press (1997)Google Scholar
- 15.Lee, H., Yoo, J., Choi, S.: Semi-supervised nonnegative matrix factorization. IEEE Signal Processing Letters 17(1), 4–7 (2010), doi:10.1109/LSP.2009.2027163Google Scholar
- 16.Shin, B., Oh, A.: Bayesian group nonnegative matrix factorization for eeg analysis. CoRR abs/1212.4347, 1–8 (2012)Google Scholar
- 17.Dohnálek, P., Gajdoš, P., Peterek, T., Penhaker, M.: Pattern recognition in EEG cognitive signals accelerated by GPU. In: Herrero, Á., Snášel, V., Abraham, A., Zelinka, I., Baruque, B., Quintián, H., Calvo, J.L., Sedano, J., Corchado, E. (eds.) Int. Joint Conf. CISIS’12-ICEUTE’12-SOCO’12. AISC, vol. 189, pp. 477–485. Springer, Heidelberg (2013)CrossRefGoogle Scholar
- 18.Prilepok, M., Platos, J., Snasel, V., Jahan, I.: Compression-based similarity in EEG signals. In: ISDA (2013)Google Scholar
- 19.Prilepok, M., Jahan, I., Snasel, V.: Electroencephalogram classification methods, Przeglad Elektrotechniczny (2013) ISSN 0033-2097, R. 89 NR 11/2013Google Scholar