Reliable classification: Learning classifiers that distinguish aleatoric and epistemic uncertainty.
Robin SengeStefan BösnerKrzysztof DembczynskiJörg HaasenritterOliver HirschNorbert Donner-BanzhoffEyke HüllermeierPublished in: Inf. Sci. (2014)
Keyphrases
- supervised learning
- multi category
- incremental learning
- classification systems
- support vector
- decision trees
- reinforcement learning
- supervised classification
- feature set
- pattern recognition
- imbalanced data sets
- accurate classification
- labeled and unlabeled data
- feature extraction
- supervised learning algorithms
- text classification
- classification rate
- classification algorithm
- nearest neighbor classifier
- classification process
- roc curve
- feature selection
- classification method
- machine learning
- active learning
- unsupervised learning
- training samples
- support vector machine svm
- class labels
- image classification
- semi supervised
- classification models
- learning process
- classifier combination
- multiple classifiers
- training set
- higher classification accuracy
- training data
- optimum path forest
- learning tasks
- probabilistic classifiers