Boosting bit rates in noninvasive EEG single-trial classifications by feature combination and multiclass paradigms.
Guido DornhegeBenjamin BlankertzGabriel CurioKlaus-Robert MüllerPublished in: IEEE Trans. Biomed. Eng. (2004)
Keyphrases
- multi class
- bit rate
- single trial
- spatial filters
- multiclass classification
- brain computer interface
- boosting algorithms
- base classifiers
- active learning
- cost sensitive
- video coding
- eeg data
- image quality
- bitstream
- event related
- feature selection
- subband
- healthy subjects
- hand movements
- support vector machine
- binary classification problems
- pairwise
- feature vectors
- binary classification
- multiclass problems
- learning algorithm
- computational complexity
- multi class classification
- coding scheme
- binary classifiers
- feature extraction
- multiresolution
- computer vision
- feature set
- image features
- image segmentation
- multiscale
- eeg signals
- pattern recognition
- learning process