k-Fold Cross-Validation Can Significantly Over-Estimate True Classification Accuracy in Common EEG-Based Passive BCI Experimental Designs: An Empirical Investigation.
Jacob WhiteSarah D. PowerPublished in: Sensors (2023)
Keyphrases
- brain computer interface
- experimental designs
- classification accuracy
- motor imagery
- signal processing
- experimental design
- healthy subjects
- brain signals
- event related potentials
- single trial
- eeg data
- eeg signals
- brain activity
- evoked potentials
- training set
- frequency ranges
- event related
- spinal cord injury
- support vector
- feature selection
- training data
- decision trees
- learning tasks
- naive bayes
- simulation platform
- graph cuts
- spatial filters
- active learning
- feature space
- brain wave
- data sets