A Method for Stopping Active Learning Based on Stabilizing Predictions and the Need for User-Adjustable Stopping.
Michael BloodgoodK. Vijay-ShankerPublished in: CoNLL (2009)
Keyphrases
- active learning
- computational complexity
- pairwise
- probabilistic model
- experimental evaluation
- computational cost
- user input
- genetic algorithm
- clustering method
- high accuracy
- relevance feedback
- cost function
- similarity measure
- support vector machine
- prior knowledge
- k means
- edge detection
- segmentation algorithm
- user interface
- detection method
- error rate