Assessing the predictive accuracy of diversity measures with domain-dependent, asymmetric misclassification costs.
Mordechai Gal-OrJerrold H. MayWilliam E. SpanglerPublished in: Inf. Fusion (2005)
Keyphrases
- predictive accuracy
- domain dependent
- diversity measures
- cost sensitive
- domain independent
- class distribution
- decision trees
- prediction accuracy
- machine learning algorithms
- multi class
- base classifiers
- classification rules
- classification models
- naive bayes
- class imbalance
- rule sets
- binary classification
- natural language
- neural network
- active learning
- clustering algorithm
- test set
- domain knowledge
- training data
- pattern recognition
- association rules
- data sets
- genetic programming
- data mining