Clean affinity matrix induced hyper-Laplacian regularization for unsupervised multi-view feature selection.
Peng SongShixuan ZhouJinshuai MuMeng DuanYanwei YuWenming ZhengPublished in: Inf. Sci. (2024)
Keyphrases
- multi view
- affinity matrix
- graph laplacian
- semi supervised
- feature selection
- labeled and unlabeled data
- kernel machines
- semi supervised learning
- unsupervised learning
- spectral clustering
- text categorization
- spectral analysis
- single view
- multi view learning
- label information
- multiple views
- unlabeled data
- supervised learning
- co training
- d objects
- depth map
- weighted graph
- labeled data
- pairwise
- euclidean space
- low rank
- three dimensional
- range images
- active learning
- feature set
- text classification
- model selection
- data clustering
- semi supervised classification
- feature extraction
- graph matching
- image segmentation
- subspace clustering
- training data
- viewpoint
- high dimensionality
- clustering method
- machine learning
- similarity measure