Hitting the target: stopping active learning at the cost-based optimum.
Zac Pullar-StreckerKatharina DostEibe FrankJörg WickerPublished in: Mach. Learn. (2024)
Keyphrases
- active learning
- active learning strategies
- cost sensitive
- cost sensitive learning
- learning strategies
- semi supervised
- high cost
- markov chain
- selective sampling
- relevance feedback
- learning algorithm
- machine learning
- neural network
- multi class
- database
- labeled data
- binary classification
- pool based active learning
- class imbalance
- global optimum
- total cost
- text classification
- data mining
- data sets