Feature Selection for Unsupervised and Supervised Inference: the Emergence of Sparsity in a Weighted-based Approach.
Lior WolfAmnon ShashuaPublished in: ICCV (2003)
Keyphrases
- feature selection
- unsupervised feature selection
- unsupervised learning
- supervised learning
- semi supervised
- unsupervised methods
- text categorization
- supervised classification
- supervised and unsupervised learning
- mutual information
- supervised feature selection
- supervised methods
- weakly supervised
- text classification
- feature selection algorithms
- bayesian inference
- irrelevant features
- machine learning
- word sense induction
- supervised training
- model selection
- semi supervised learning
- bayesian networks
- learning algorithm
- feature extraction
- multi class
- support vector
- classification accuracy
- feature set
- discriminant projection
- feature space
- unsupervised manner
- discriminative features
- knn
- support vector machine
- belief networks
- feature subset
- dimensionality reduction
- active learning
- probabilistic inference
- high dimensional
- pairwise
- training set
- high dimensionality
- sparse representation
- feature selection and classification
- unlabeled data
- graphical models