Investigation on the stability of SMOTE-based oversampling techniques in software defect prediction.
Shuo FengJacky KeungXiao YuYan XiaoMiao ZhangPublished in: Inf. Softw. Technol. (2021)
Keyphrases
- software defect prediction
- class imbalance
- imbalanced data
- class distribution
- minority class
- cost sensitive
- active learning
- concept drift
- sampling methods
- cost sensitive learning
- high dimensionality
- feature selection
- ensemble learning
- training set
- random forest
- decision boundary
- misclassification costs
- decision trees
- training data
- data streams
- feature vectors
- prediction accuracy
- training examples
- training samples
- dimensionality reduction