Optimised probabilistic active learning (OPAL) - For fast, non-myopic, cost-sensitive active classification.
Georg KremplDaniel KottkeVincent LemairePublished in: Mach. Learn. (2015)
Keyphrases
- cost sensitive
- active learning
- cost sensitive classification
- cost sensitive learning
- uncertainty sampling
- class imbalance
- confidence weighted
- misclassification costs
- multi class
- binary classification
- class distribution
- machine learning
- random sampling
- labeled data
- supervised learning
- training set
- semi supervised
- support vector machine
- training examples
- learning algorithm
- bayesian networks
- probability estimates
- learning process
- error rate
- classification algorithm
- unlabeled data
- boosting algorithms
- class dependent
- decision trees
- active learning strategies
- base classifiers
- support vector
- naive bayes
- classification accuracy
- multi label
- cross validation
- class labels
- benchmark datasets
- image classification
- data mining