CDSMOTE: class decomposition and synthetic minority class oversampling technique for imbalanced-data classification.
Eyad ElyanCarlos Francisco Moreno-GarcíaChrisina JaynePublished in: Neural Comput. Appl. (2021)
Keyphrases
- imbalanced data classification
- minority class
- class imbalance
- majority class
- class distribution
- cost sensitive learning
- classification error
- support vector machine
- feature selection
- training dataset
- decision boundary
- nearest neighbour
- original data
- sampling methods
- ensemble learning
- training data
- cost sensitive
- active learning
- multi class
- training set
- feature vectors
- machine learning
- feature extraction
- data streams
- random sampling
- base classifiers
- test set
- non stationary
- feature set