Interpretable Discriminative Dimensionality Reduction and Feature Selection on the Manifold.
Babak HosseiniBarbara HammerPublished in: ECML/PKDD (1) (2019)
Keyphrases
- dimensionality reduction
- feature selection
- manifold learning
- low dimensional
- lower dimensional
- diffusion maps
- nonlinear dimensionality reduction
- feature extraction
- class separability
- manifold structure
- feature space
- locally linear embedding
- class discrimination
- latent space
- high dimensionality
- discriminative features
- unsupervised learning
- graph embedding
- high dimensional
- high dimensional data
- locality preserving
- nonlinear manifold
- data representation
- high dimension
- text categorization
- linear discriminant analysis
- discriminant information
- subspace learning
- machine learning
- underlying manifold
- classification accuracy
- text classification
- dimension reduction
- embedding space
- random projections
- principal component analysis
- dimensionality reduction methods
- feature selection algorithms
- feature set
- linear dimensionality reduction
- unsupervised feature selection
- support vector machine
- informative features
- principal components
- euclidean space
- locality preserving projections
- multi class
- linear subspace
- pattern recognition
- sparse representation
- support vector
- kernel pca
- data points
- semi supervised
- selected features
- euclidean distance