Using Higher Order Nonlinear Operators for SVM Classification of EEG Data
AbstractBrain-Computer Interface (BCI) systems require application of complex analysis, signal processing, denoising, feature extraction, dimensionality reduction and classification methods on acquired raw electroencephalogram (EEG) data to allow for useful operation. In this paper, we consider application of nonlinear operators such as Taeger-Kaiser Energy Operator (TKEO) and its multiple generalizations on the EEG signals and evaluate the efficiency of the operators using a Support Vector Machine (SVM) classifier with linear kernel. We propose a new generalization of TKEO, called Homogeneous Multivariate Polynomial Operator (HMPO), and compare the efficiency of the 3rd order HMPO with other nonlinear operators. Experimental results show that the 3rd order HMPO operator allows for better identification of significant features representing slow cortical potentials in the EEG data. Ill. 1, bibl. 18, tabl. 1 (in English; abstracts in English and Lithuanian).
Authors retain copyright and grant the journal the right of the first publication with the paper simultaneously licensed under the Creative Commons Attribution 4.0 (CC BY 4.0) licence.
Authors are allowed to enter into separate, additional contractual arrangements for the non-exclusive distribution of the paper published in the journal with an acknowledgement of the initial publication in the journal.
Copyright terms are indicated in the Republic of Lithuania Law on Copyright and Related Rights, Articles 4-37.