CHSMOTE: Convex hull-based synthetic minority oversampling technique for alleviating the class imbalance problem.
Xiaohan YuanShuyu ChenHan ZhouChuan SunYuwen LuPublished in: Inf. Sci. (2023)
Keyphrases
- convex hull
- class imbalance
- minority class
- class distribution
- majority class
- class imbalanced
- active learning
- cost sensitive
- training samples
- cost sensitive learning
- imbalanced datasets
- concept drift
- imbalanced data
- misclassification costs
- training data
- high dimensionality
- feature selection
- sampling methods
- hyperplane
- test set
- imbalanced class distribution
- image processing
- convex polyhedra
- training set
- training examples
- unlabeled data
- decision boundary
- nearest neighbour
- classification error
- data mining
- class labels
- computer vision