A new approach to dimensionality reduction based on locality preserving LDA.
Di ZhangJiazhong HePublished in: FSKD (2013)
Keyphrases
- dimensionality reduction
- locality preserving
- linear discriminant analysis
- locality preserving projections
- manifold learning
- kernel trick
- high dimensional data
- principal component analysis
- discriminant analysis
- dimensionality reduction methods
- pattern recognition
- low dimensional
- high dimensional
- latent dirichlet allocation
- lower dimensional
- data representation
- feature extraction
- feature space
- feature selection
- subspace learning
- euclidean distance
- manifold structure
- embedding space
- random projections
- nonlinear dimensionality reduction
- principal components
- dimension reduction
- input space
- face recognition
- data points
- singular value decomposition
- metric learning
- data sets
- discriminative information
- graph embedding
- high dimensionality
- latent space
- locally linear embedding
- topic models
- knn
- principal components analysis
- pairwise
- data analysis
- latent semantic indexing
- discriminant information