Improving financial bankruptcy prediction in a highly imbalanced class distribution using oversampling and ensemble learning: a case from the Spanish market.
Hossam FarisRuba AbukhurmaWaref AlmanaseerMohammed SaadehAntonio Mora GarcíaPedro A. CastilloIbrahim AljarahPublished in: Prog. Artif. Intell. (2020)
Keyphrases
- class distribution
- ensemble learning
- highly imbalanced
- base classifiers
- class imbalance
- concept drift
- financial data
- imbalanced data
- unlabeled data
- ensemble methods
- minority class
- cost sensitive
- misclassification costs
- multi class
- training data
- training set
- stock market
- cost sensitive learning
- decision trees
- ensemble classifier
- test set
- classification error
- data sets
- generalization ability
- test data
- training samples
- random forest
- labeled data
- class labels
- data streams
- classification algorithm
- semi supervised learning
- semi supervised
- remote sensing
- non stationary
- training examples
- decision boundary
- change detection