Mutual information criterion for feature selection from incomplete data.
Wenbin QianWenhao ShuPublished in: Neurocomputing (2015)
Keyphrases
- mutual information
- incomplete data
- feature selection
- information theoretic criterion
- selection criterion
- learning bayesian networks
- missing values
- conditional mutual information
- criterion function
- information theoretic
- normalized mutual information
- irrelevant attributes
- missing data
- bayesian networks
- information gain
- bayes classifier
- text categorization
- medical image registration
- incomplete data sets
- em algorithm
- dimensionality reduction
- image registration
- feature set
- similarity measure
- feature space
- feature selection algorithms
- support vector
- multi class
- text classification
- typical testors
- supervised feature selection
- machine learning
- classification accuracy
- feature extraction
- support vector machine
- selecting relevant features
- data sets
- variational bayesian
- feature weighting
- multi task
- classification models
- database