A multi-criteria meta-learning method to select under-sampling algorithms for imbalanced datasets.
Romero F. A. B. de MoraisPéricles B. C. de MirandaRicardo M. A. SilvaPublished in: ESANN (2017)
Keyphrases
- multi criteria
- meta learning
- learning algorithm
- artificial intelligence
- data sets
- objective function
- sampling methods
- prediction accuracy
- pairwise
- classification accuracy
- multi class
- decision makers
- image classification
- machine learning algorithms
- prior knowledge
- benchmark datasets
- similarity measure
- nearest neighbour
- decision making