Ensemble method using real images, metadata and synthetic images for control of class imbalance in classification.
Rogers AlooAtsuko MutohKoichi MoriyamaTohgoroh MatsuiNobuhiro InuzukaPublished in: Artif. Life Robotics (2022)
Keyphrases
- class imbalance
- ensemble methods
- image classification
- class distribution
- machine learning methods
- cost sensitive
- decision trees
- generalization ability
- active learning
- benchmark datasets
- input image
- pattern recognition
- high dimensionality
- feature selection
- prediction accuracy
- small number
- image features
- feature space
- machine learning
- cross validation
- image registration
- classification models
- random forests
- classification accuracy
- minority class
- text classification
- misclassification costs
- ensemble classifier
- training set
- ensemble learning
- random forest
- support vector machine
- multi class
- class labels
- naive bayes