A Bregman-proximal point algorithm for robust non-negative matrix factorization with possible missing values and outliers - application to gene expression analysis.
Stéphane ChrétienChristophe GuyeuxBastien ConesaRégis Delage-MourouxMichèle JouvenotPhilippe HuetzFrançoise DescôtesPublished in: BMC Bioinform. (2016)
Keyphrases
- missing values
- negative matrix factorization
- missing data
- gene expression analysis
- matrix factorization
- sparse representation
- high dimensional data
- machine learning
- nonnegative matrix factorization
- neural network
- denoising
- image restoration
- data sets
- natural language processing
- similarity measure
- document clustering
- total variation
- computer vision