A new Bayesian approach to nonnegative matrix factorization: Uniqueness and model order selection.
Reinhard SchachtnerGerhard PöppelAna Maria ToméCarlos García PuntonetElmar Wolfgang LangPublished in: Neurocomputing (2014)
Keyphrases
- nonnegative matrix factorization
- model order selection
- parameter tuning
- probabilistic latent semantic indexing
- data representation
- matrix factorization
- negative matrix factorization
- least squares
- active set
- nonnegative matrix
- principal component analysis
- spectral clustering
- original data
- document clustering
- objective function
- clustering algorithm
- raw data
- missing data
- image retrieval
- pattern recognition