Bayesian approach to feature selection and parameter tuning for support vector machine classifiers.
Carl GoldAlex HolubPeter SollichPublished in: Neural Networks (2005)
Keyphrases
- parameter tuning
- support vector machine classifiers
- feature selection
- support vector machine
- ink bleed
- parameter settings
- text categorization
- model selection
- multi class
- classification models
- feature space
- feature set
- selected features
- classification accuracy
- support vector
- feature selection algorithms
- text classification
- higher order
- machine learning
- feature subset
- data sets
- svm classifier
- evolutionary algorithm
- feature vectors
- bayes classifier
- search algorithm