Using a penalized maximum likelihood model for feature selection.
Amir JalaliradTjalling J. TjalkensPublished in: MLSP (2015)
Keyphrases
- likelihood model
- feature selection
- probabilistic model
- text categorization
- model selection
- support vector
- least squares
- mutual information
- maximum number
- feature extraction
- classification accuracy
- maximum likelihood
- text classification
- loss function
- database
- feature selection and classification
- variable selection
- discriminative features
- irrelevant features
- feature space
- machine learning
- minimum redundancy
- feature set
- unsupervised feature selection
- feature selection algorithms
- selected features
- small sample
- informative features
- forward selection
- conditional mutual information
- high dimensionality
- bayes classifier
- multi task
- information gain
- microarray data
- multi class
- preprocessing
- data sets