A-SMOTE: A New Preprocessing Approach for Highly Imbalanced Datasets by Improving SMOTE.
Ahmed Saad HusseinTianrui LiChubato Wondaferaw YohanneseKamal BashirPublished in: Int. J. Comput. Intell. Syst. (2019)
Keyphrases
- highly imbalanced
- class distribution
- imbalanced data
- imbalanced datasets
- preprocessing
- data sets
- class imbalance
- class imbalanced
- cost sensitive
- sampling methods
- test set
- training set
- minority class
- training samples
- misclassification costs
- random forest
- linear regression
- test data
- training data
- class labels
- machine learning
- cost sensitive learning
- decision trees
- feature selection
- database
- machine learning algorithms
- error rate