An active learning budget-based oversampling approach for partially labeled multi-class imbalanced data streams.
Gabriel AguiarAlberto CanoPublished in: SAC (2023)
Keyphrases
- active learning
- class imbalance
- class imbalanced
- data streams
- partially labeled
- semi supervised
- semi supervised learning
- concept drift
- minority class
- unlabeled data
- sliding window
- random sampling
- labeled data
- machine learning
- learning algorithm
- data sets
- supervised learning
- sampling methods
- binary classification
- text classification
- feature selection
- prior knowledge
- generalization error
- class distribution
- learning process
- feature extraction
- cost sensitive
- transfer learning
- data distribution
- prediction accuracy
- training examples