Original title: Classification Methods for Brain-Computer Interface
Authors: Bobrov, P. ; Frolov, A. A. ; Húsek, Dušan
Document type: Papers
Conference/Event: WOFEX 2011. Annual Workshop /9./, Ostrava (CZ), 2011-09-08 / 2011-09-09
Year: 2011
Language: eng
Abstract: The performance of four classifiers for Brain Computer Interface (BCI) systems based on multichannel EEG recordings is tested in this work. The classifiers are designed to distinguish EEG patterns corresponding to performance of several mental tasks. It is shown that relatively simple classifiers based on the Bayesian approach are comparable in classification accuracy with more sophisticated classifiers based on Common Spatial Patterns and Common Tensor Discriminant Analysis
Keywords: Bayesian classifier; BCI; brain computer interface; classification accuracy; common spatial patterns; common tensor discriminant analysis; CSP; CTDA; EEG
Project no.: CEZ:AV0Z10300504 (CEP)
Host item entry: WOFEX 2011, ISBN 978-80-248-2449-9

Institution: Institute of Computer Science AS ČR (web)
Document availability information: Fulltext is available on demand via the digital repository of the Academy of Sciences.
Original record: http://hdl.handle.net/11104/0199687

Permalink: http://www.nusl.cz/ntk/nusl-55913


The record appears in these collections:
Research > Institutes ASCR > Institute of Computer Science
Conference materials > Papers