A Rule-Based Scheme for Filtering Examples from Majority Class in an Imbalanced Training Set.
Jamshid DehmeshkiMustafa KaraköyManlio Valdivieso CasiquePublished in: MLDM (2003)
Keyphrases
- majority class
- minority class
- class imbalance
- class distribution
- training set
- imbalanced data
- imbalanced datasets
- training examples
- active learning
- cost sensitive
- decision boundary
- classification error
- nearest neighbour
- training data
- support vector machine
- misclassification costs
- test set
- binary classification
- training dataset
- cost sensitive learning
- data sets
- unlabeled data
- class labels
- attribute selection
- classification accuracy
- original data
- concept drift
- high dimensionality
- cross validation
- training samples
- decision trees
- sampling methods
- rare class
- linear regression
- ensemble learning
- test data
- face images
- supervised learning
- nearest neighbor
- learning algorithm