Manifold embedded instance selection to suppress negative transfer in motor imagery-based brain-computer interface.
Zilin LiangZheng ZhengWeihai ChenZhongcai PeiJianhua WangJianer ChenPublished in: Biomed. Signal Process. Control. (2024)
Keyphrases
- motor imagery
- instance selection
- brain computer interface
- feature and instance selection
- data reduction
- nearest neighbor
- text classification
- multiple instance learning
- classification accuracy
- semi supervised learning
- supervised learning
- positive and negative
- multi class
- transfer learning
- eeg signals
- evaluation criteria
- knowledge discovery and data mining
- eeg data
- healthy subjects
- data analysis
- blind source separation
- high dimensional
- concept drift
- pattern recognition
- model selection
- low dimensional
- semi supervised
- information extraction
- preprocessing