Cross project defect prediction using class distribution estimation and oversampling.
Nachai LimsetthoKwabena Ebo BenninJacky W. KeungHideaki HataKenichi MatsumotoPublished in: Inf. Softw. Technol. (2018)
Keyphrases
- class distribution
- class imbalance
- defect prediction
- majority class
- minority class
- software projects
- cost sensitive
- training data
- training set
- imbalanced data
- misclassification costs
- concept drift
- cost sensitive learning
- highly skewed
- training samples
- test set
- highly imbalanced
- case study
- base classifiers
- project management
- sampling methods
- training examples
- active learning
- high dimensionality
- class labels
- open source
- bayesian networks
- machine learning