A multi-schematic classifier-independent oversampling approach for imbalanced datasets.
Saptarshi BejKristian SchultzPrashant SrivastavaMarkus WolfienOlaf WolkenhauerPublished in: CoRR (2021)
Keyphrases
- imbalanced datasets
- minority class
- class imbalance
- majority class
- cost sensitive learning
- class distribution
- highly skewed
- decision trees
- sampling methods
- learning from imbalanced data
- imbalanced data
- training dataset
- classification error
- feature selection algorithms
- training data
- active learning
- decision boundary
- support vector machine
- feature selection
- nearest neighbour
- cost sensitive
- ensemble learning
- ensemble methods
- training set
- rare class
- missing values
- misclassification costs
- class labels
- learning algorithm
- test set
- binary classification
- training examples
- original data
- feature extraction
- feature space
- concept drift
- training samples
- probability estimation
- classification trees