Nonnegative Matrix Factorization and Random Forest for Classification of Heart Sound Recordings in the Spectral Domain.
Christoph Hoog AntinkJulian BeckerSteffen LeonhardtMarian WalterPublished in: CinC (2016)
Keyphrases
- random forest
- nonnegative matrix factorization
- decision trees
- feature set
- spectral domain
- ensemble methods
- pattern recognition
- classification accuracy
- frequency domain
- matrix factorization
- support vector
- multi label
- image classification
- machine learning
- benchmark datasets
- feature extraction
- feature selection
- training samples
- least squares
- feature space
- data representation
- multi class
- class labels
- machine learning methods
- original data
- spectral clustering
- spatial domain
- training data
- similarity measure
- data mining