Influence of minority class instance types on SMOTE imbalanced data oversampling.
Przemyslaw SkryjomskiBartosz KrawczykPublished in: LIDTA@PKDD/ECML (2017)
Keyphrases
- imbalanced data
- minority class
- class imbalance
- majority class
- class distribution
- classification error
- imbalanced datasets
- imbalanced data sets
- cost sensitive learning
- sampling methods
- support vector machine
- nearest neighbour
- decision boundary
- training dataset
- class imbalanced
- rare events
- original data
- training data
- linear regression
- training set
- highly skewed
- cost sensitive
- feature selection
- decision trees
- classification models
- ensemble methods
- active learning
- data sets
- ensemble learning
- machine learning
- svm classifier
- concept drift
- high dimensionality
- ensemble classifier
- pattern classification
- feature selection algorithms