Clustering underlying stock trends via non-negative matrix factorization.
Andrea PazienzaSabrina Francesca PellegrinoStefano FerilliFloriana EspositoPublished in: MIDAS@PKDD/ECML (2016)
Keyphrases
- negative matrix factorization
- document clustering
- spectral clustering
- nonnegative matrix factorization
- probabilistic latent semantic analysis
- clustering method
- clustering algorithm
- k means
- unsupervised feature selection
- sparse representation
- principal component analysis
- matrix factorization
- tensor factorization
- data clustering
- constrained least squares
- latent semantic space
- cluster analysis
- data points
- pairwise
- computer vision
- document collections
- generative model
- dimensionality reduction
- knowledge base