Seizure prediction with spectral power of time/space-differential EEG signals using cost-sensitive support vector machine.
Yun S. ParkTheoden I. NetoffKeshab K. ParhiPublished in: ICASSP (2010)
Keyphrases
- cost sensitive
- eeg signals
- support vector machine
- multi class
- epileptic seizures
- cost sensitive learning
- misclassification costs
- cost sensitive classification
- binary classification
- brain activity
- signal processing
- eeg data
- prediction accuracy
- support vector
- brain computer interface
- svm classifier
- feature selection
- class distribution
- active learning
- class imbalance
- machine learning
- decision boundary
- support vector machine svm
- feature space
- naive bayes
- k nearest neighbor
- extracted features
- non stationary
- rough sets
- boosting algorithms
- feature vectors
- training data
- image segmentation
- data sets
- cost sensitive boosting