AnchorAL: Computationally Efficient Active Learning for Large and Imbalanced Datasets.
Pietro LesciAndreas VlachosPublished in: NAACL-HLT (2024)
Keyphrases
- imbalanced datasets
- computationally efficient
- active learning
- class imbalance
- rare class
- cost sensitive learning
- imbalanced class distribution
- learning from imbalanced data
- cost sensitive
- class distribution
- sampling methods
- decision trees
- semi supervised
- random sampling
- binary classification
- minority class
- learning algorithm
- machine learning
- feature selection algorithms
- training set
- ensemble methods
- semi supervised learning
- supervised learning
- imbalanced data
- misclassification costs
- labeled data
- small number
- support vector machine
- learning process