A supervised class-preserving Laplacian eigenmaps for dimensionality reduction.
Ning WangJinyan LiuTingquan DengPublished in: ICNC-FSKD (2016)
Keyphrases
- laplacian eigenmaps
- dimensionality reduction
- label information
- manifold learning
- nonlinear dimensionality reduction
- kernel pca
- low dimensional
- locally linear embedding
- high dimensional data
- dimensionality reduction methods
- principal component analysis
- semi supervised
- unsupervised learning
- high dimensional
- feature selection
- subspace learning
- high dimensionality
- random projections
- feature space
- pattern recognition
- linear discriminant analysis
- data points
- feature extraction
- image processing
- learning algorithm
- spectral clustering
- principal components
- kernel methods
- euclidean space
- preprocessing step
- multi dimensional
- graph laplacian