VFC-SMOTE: very fast continuous synthetic minority oversampling for evolving data streams.
Alessio BernardoEmanuele Della VallePublished in: Data Min. Knowl. Discov. (2021)
Keyphrases
- evolving data streams
- minority class
- class imbalance
- concept drift
- class distribution
- class imbalanced
- majority class
- contour extraction
- b spline
- data streams
- streaming data
- gradient vector flow
- imbalanced data
- stream mining
- random forests
- cost sensitive learning
- classification error
- data stream mining
- decision boundary
- nearest neighbour
- support vector machine
- active learning
- sampling methods
- active contour model
- active contours
- training set
- perceptual grouping
- cost sensitive
- change detection
- least squares
- training data
- decision trees
- classification algorithm
- sliding window
- non stationary
- anomaly detection