DocumentCode :
3041425
Title :
A New Feature Dimensionally Reduction Approach Based on General Tensor Discriminant Analysis in EEG Signal Classification
Author :
Nasehi, Saadat ; Pourghassem, Hossein
Author_Institution :
Dept. of Electr. Eng., Islamic Azad Univ., Isfahan, Iran
fYear :
2011
fDate :
14-17 Dec. 2011
Firstpage :
188
Lastpage :
191
Abstract :
Feature selection from electroencephalogram (EEG) signals is important steps in BCI and medicine application. In this paper, a feature dimensionally reduction approach based on general tensor discriminant analysis (GTDA) is proposed. In this approach, EEG signal epochs are decomposed as spectral, spatial and temporal domain by Gabor functions as third order tensors. Then, projection vectors are extracted from tensor-represented EEG signals by GTDA. In this approach, the discriminative information in the training tensors is preserved that is a benefit in comparison with common feature space reduction approaches such as linear discriminant analysis (LDA) and principal component analysis (PCA). The proposed approach is evaluated to classify three mental tasks. The results indicate the improvement of classification performance in comparison with current methods.
Keywords :
Gabor filters; brain-computer interfaces; electroencephalography; feature extraction; functions; medical signal processing; tensors; vectors; BCI; EEG signal classification; GTDA; Gabor functions; discriminative information; electroencephalogram signals; feature dimensionally reduction approach; feature selection; general tensor discriminant analysis; mental tasks; projection vector; spatial domain; spectral domain; temporal domain; tensor represented EEG signal epoch; third order tensor; training tensors; Accuracy; Electroencephalography; Feature extraction; Pattern classification; Principal component analysis; Tensile stress; Training; EEG; Gabor functions; General Tensor Discriminant Analysis (GTDA); ICA; LDA; PCA; feature extraction;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Computation and Bio-Medical Instrumentation (ICBMI), 2011 International Conference on
Conference_Location :
Wuhan, Hubei
Print_ISBN :
978-1-4577-1152-7
Type :
conf
DOI :
10.1109/ICBMI.2011.32
Filename :
6131743
Link To Document :
بازگشت