Using independent component analysis to obtain feature space for reliable ECG Arrhythmia classification.
Mohammad SarfrazAteeq Ahmed KhanFrancis F. LiPublished in: BIBM (2014)
Keyphrases
- feature space
- classification accuracy
- heart rate variability
- feature vectors
- pattern recognition
- mit bih arrhythmia database
- training set
- classification systems
- high dimensionality
- support vector machine
- training samples
- feature set
- high dimensional
- support vector
- feature selection
- ecg signals
- kernel function
- preprocessing
- automatic classification
- svm classifier
- machine learning
- beat classification
- high dimension
- decision rules
- supervised learning
- feature extraction
- image classification
- model selection
- machine learning algorithms
- classification method
- kernel methods
- decision trees
- dimension reduction
- pattern classification
- principal component analysis
- support vector machine svm
- active learning
- learning algorithm
- high dimensional feature space
- image retrieval
- spectral analysis
- image representation
- classification scheme
- hyperplane
- linear discriminant analysis
- classification rules
- input data
- text classification
- class labels
- neural network