An Empirical Study of the Behavior of Classifiers on Imbalanced and Overlapped Data Sets.
Vicente GarcíaJosé Salvador SánchezRamón Alberto MollinedaPublished in: CIARP (2007)
Keyphrases
- data sets
- imbalanced data sets
- training set
- training data
- benchmark data sets
- decision trees
- support vector
- imbalanced data
- class labels
- linear classifiers
- machine learning algorithms
- binary classification problems
- classification error
- bayesian network classifiers
- classification trees
- database
- training samples
- real world
- class distribution
- majority voting
- minority class
- high dimensional data
- unseen data
- highly imbalanced