Classifier Selection for Highly Imbalanced Data Streams with Minority Driven Ensemble.
Pawel ZyblewskiPawel KsieniewiczMichal WozniakPublished in: ICAISC (1) (2019)
Keyphrases
- highly imbalanced
- class distribution
- data streams
- imbalanced data
- concept drift
- ensemble classifier
- training data
- ensemble learning
- training set
- data sets
- base classifiers
- class imbalance
- sliding window
- decision trees
- feature selection
- minority class
- support vector machine
- class labels
- classifier ensemble
- ensemble methods
- change detection
- classification algorithm
- training examples
- cost sensitive
- training samples
- random forest
- multi class
- random forests
- classification models
- cost sensitive learning
- feature space
- unlabeled data
- svm classifier
- decision boundary
- non stationary
- misclassification costs
- classification accuracy
- test data
- weak classifiers
- machine learning methods
- data distribution
- support vector
- learning algorithm
- machine learning