Structure constrained semi-nonnegative matrix factorization for EEG-based motor imagery classification.
Na LuTengfei LiJinjin PanXiaodong RenZuren FengHongyu MiaoPublished in: Comput. Biol. Medicine (2015)
Keyphrases
- motor imagery
- nonnegative matrix factorization
- eeg signals
- brain computer interface
- eeg data
- matrix factorization
- negative matrix factorization
- least squares
- machine learning
- pattern recognition
- healthy subjects
- classification accuracy
- spinal cord injury
- data mining
- brain activity
- blind source separation
- data representation
- bag of words
- natural language processing
- training set
- support vector
- feature selection