An overview of ensemble methods for binary classifiers in multi-class problems: Experimental study on one-vs-one and one-vs-all schemes.
Mikel GalarAlberto FernándezEdurne Barrenechea TartasHumberto Bustince SolaFrancisco HerreraPublished in: Pattern Recognit. (2011)
Keyphrases
- binary classifiers
- ensemble methods
- binary classification
- multi class problems
- prediction accuracy
- multi class
- multi class classification
- binary classification problems
- decision trees
- multiclass classification
- benchmark datasets
- random forests
- ensemble learning
- machine learning methods
- base classifiers
- generalization ability
- misclassification costs
- majority voting
- cross validation
- base learners
- learning problems
- training samples
- feature selection
- machine learning
- cost sensitive learning
- support vector machine