Divide-and-Conquer Anchoring for Near-Separable Nonnegative Matrix Factorization and Completion in High Dimensions.
Tianyi ZhouWei BianDacheng TaoPublished in: ICDM (2013)
Keyphrases
- nonnegative matrix factorization
- high dimensions
- high dimensional data
- matrix factorization
- high dimensional
- negative matrix factorization
- data representation
- least squares
- original data
- spectral clustering
- poor quality
- high dimensional spaces
- document clustering
- data sets
- principal component analysis
- image data
- data points
- feature extraction
- image processing