The mean subjective utility score, a novel metric for cost-sensitive classifier evaluation.
Ross A. McDonaldPublished in: Pattern Recognit. Lett. (2006)
Keyphrases
- cost sensitive
- cost sensitive classification
- cost sensitive learning
- probability estimates
- multi class
- misclassification costs
- class probabilities
- binary classification
- class dependent
- support vector machine
- active learning
- binary classifiers
- class distribution
- naive bayes
- confidence weighted
- class imbalance
- linear classifiers
- training data
- feature selection
- neural network
- error rate
- training examples
- training samples
- training set
- decision trees
- data sets