Feature selection and multi-kernel learning for adaptive graph regularized nonnegative matrix factorization.
Jim Jing-Yan WangJianhua Z. HuangYijun SunXin GaoPublished in: Expert Syst. Appl. (2015)
Keyphrases
- kernel learning
- nonnegative matrix factorization
- sparsity constraints
- feature selection
- least squares
- multiple kernel learning
- semi supervised
- pairwise constraints
- spectral clustering
- kernel matrix
- data representation
- kernel function
- matrix factorization
- support vector regression
- negative matrix factorization
- kernel methods
- objective function
- learning tasks
- document clustering
- distance metric
- dimensionality reduction
- support vector
- principal component analysis
- data mining
- pairwise
- low rank
- linear combination
- model selection
- text classification
- image retrieval