A Filter-Wrapper Method to Select Variables for the Naive Bayes Classifier Based on Credal Decision Trees.
Joaquín AbellánAndrés R. MasegosaPublished in: Int. J. Uncertain. Fuzziness Knowl. Based Syst. (2009)
Keyphrases
- naive bayes
- wrapper method
- decision trees
- feature selection
- text classification
- logistic regression
- support vector machine
- classification accuracy
- probability estimation
- naive bayes classifier
- text categorization
- information gain
- cost sensitive
- bayesian classifier
- training data
- gene selection
- leaf nodes
- machine learning
- base classifiers
- model averaging
- bayesian network classifiers
- probability estimates
- variable selection
- machine learning algorithms
- conditional independence assumption
- boosted decision trees
- neural network
- averaged one dependence estimators
- random forest
- training set
- ensemble methods
- combining classifiers
- conditional probabilities
- probabilistic classifiers
- random variables
- support vector machine svm
- bayesian networks