Constrained Log-Likelihood-Based Semi-supervised Linear Discriminant Analysis.
Marco LoogAre Charles JensenPublished in: SSPR/SPR (2012)
Keyphrases
- linear discriminant analysis
- log likelihood
- semi supervised
- discriminant analysis
- face recognition
- dimensionality reduction
- principal component analysis
- dimension reduction
- density estimation
- information theoretic
- maximum likelihood
- high dimensional data
- semi supervised learning
- em algorithm
- scoring function
- feature space
- null space
- feature extraction
- support vector
- subspace learning
- supervised learning
- pairwise
- labeled data
- support vector machine svm
- unlabeled data
- closed form
- nearest neighbor
- data streams
- scatter matrices
- machine learning
- unsupervised learning
- generative model
- classification accuracy
- active learning
- high dimensional
- pattern recognition
- learning algorithm