A Consensus Approach for Combining Multiple Classifiers in Cost-Sensitive Bankruptcy Prediction.
Ning ChenBernardete RibeiroPublished in: ICANNGA (2013)
Keyphrases
- cost sensitive
- combining multiple
- bankruptcy prediction
- naive bayes
- cluster ensemble
- fusion methods
- credit card
- fraud detection
- boosting algorithms
- multi class
- misclassification costs
- consensus clustering
- cost sensitive learning
- binary classification
- decision trees
- binary classifiers
- class distribution
- financial data
- learning vector quantization
- cost sensitive classification
- clustering ensemble
- literature review
- classification accuracy
- training examples
- text classification
- active learning
- support vector
- linear classifiers
- training data
- class imbalance
- feature set
- feature selection
- training set
- base classifiers
- bayesian networks
- probability estimation
- classification algorithm
- k means
- test set
- multi label