Assessing the Costs of Sampling Methods in Active Learning for Annotation.
Robbie HaertelEric K. RinggerKevin D. SeppiJames L. CarrollPeter McClanahanPublished in: ACL (2) (2008)
Keyphrases
- worst case
- active learning
- sampling methods
- random sampling
- class imbalance
- cost sensitive learning
- imbalanced datasets
- misclassification costs
- sampling algorithm
- stratified sampling
- semi supervised
- training examples
- imbalanced data
- machine learning
- learning algorithm
- sample size
- cost sensitive
- supervised learning
- training set
- semi supervised learning
- learning process
- pairwise
- unlabeled data
- labeled data
- minority class
- reinforcement learning