Maximizing classifier utility when training data is costly.
Gary M. WeissYe TianPublished in: SIGKDD Explor. (2006)
Keyphrases
- training data
- training set
- training examples
- decision trees
- training samples
- training instances
- training process
- labeled training data
- learning algorithm
- training dataset
- class labels
- test data
- supervised learning
- classification models
- labeled data
- support vector machine
- data sets
- sufficient training data
- classification accuracy
- manually labeled training data
- error prone
- utility function
- test set
- semi supervised learning
- support vector
- feature selection
- prior knowledge
- feature space
- classification trees
- training and test data
- unlabeled data