Improvements in Approximation Performance and Parallelization of Nonnegative Matrix Factorization with Newton Iteration.
Rade KutilMarkus FlatzMarián VajtersicPublished in: HPCS (2017)
Keyphrases
- nonnegative matrix factorization
- interior point
- newton method
- matrix decomposition
- least squares
- negative matrix factorization
- probabilistic latent semantic indexing
- matrix factorization
- data representation
- active set
- nonnegative matrix
- objective function
- spectral clustering
- approximation algorithms
- principal component analysis
- original data
- document clustering
- natural language processing
- knn
- knowledge discovery