Sources identification using shifted non-negative matrix factorization combined with semi-supervised clustering.
Filip L. IlievValentin G. StanevVelimir V. VesselinovBoian S. AlexandrovPublished in: CoRR (2016)
Keyphrases
- negative matrix factorization
- semi supervised clustering
- nonnegative matrix factorization
- document clustering
- matrix factorization
- pairwise constraints
- sparse representation
- principal component analysis
- semi supervised
- metric learning
- spectral clustering
- data representation
- background knowledge
- k means
- supervised learning
- text mining
- labeled data
- information retrieval
- knn
- active learning
- pairwise
- pattern recognition