An Empirical Study of Bagging Predictors for Imbalanced Data with Different Levels of Class Distribution.
Guohua LiangXingquan ZhuChengqi ZhangPublished in: Australasian Conference on Artificial Intelligence (2011)
Keyphrases
- imbalanced data
- class distribution
- class imbalance
- highly imbalanced
- imbalanced datasets
- misclassification costs
- cost sensitive
- highly skewed
- training data
- training set
- test set
- majority class
- linear regression
- test data
- minority class
- unlabeled data
- training samples
- imbalanced data sets
- ensemble methods
- concept drift
- class labels
- active learning
- face recognition
- decision trees
- base classifiers
- benchmark datasets
- prediction accuracy
- training examples
- support vector machine