An asymptotic approximation for EPMC in linear discriminant analysis based on two-step monotone missing samples.
Nobumichi ShutohMasashi HyodoTakashi SeoPublished in: J. Multivar. Anal. (2011)
Keyphrases
- linear discriminant analysis
- scatter matrix
- discriminant analysis
- face recognition
- dimensionality reduction
- principal component analysis
- dimension reduction
- feature extraction
- support vector
- discriminant features
- feature space
- null space
- high dimensional data
- fisher criterion
- data sets
- discriminant information
- small sample size
- feature selection
- training samples
- missing data
- discriminative information
- high dimensional
- nearest neighbor
- upper bound
- pattern recognition
- class separability
- linear discriminant
- subspace analysis
- generalized discriminant analysis
- scatter matrices
- locality preserving projections
- missing values
- sample size
- low dimensional
- feature vectors