Some features speak loud, but together they all speak louder: A study on the correlation between classification error and feature usage in decision-tree classification ensembles.
Bárbara CervantesRaúl MonroyMiguel Angel Medina-PérezMiguel González-MendozaJose Emmanuel Ramirez-MarquezPublished in: Eng. Appl. Artif. Intell. (2018)
Keyphrases
- decision trees
- classification error
- feature set
- feature vectors
- class labels
- classification models
- decision rules
- base classifiers
- classification accuracy
- training set
- feature analysis
- feature extraction
- feature space
- weak classifiers
- training samples
- image features
- feature subset
- feature selection
- naive bayes
- training data
- text classification
- ensemble methods
- support vector
- machine learning algorithms
- benchmark datasets
- svm classifier
- prior knowledge
- pattern recognition
- information gain
- generalization error
- machine learning
- model selection
- image classification
- supervised learning
- face recognition