Cost-Sensitive Perceptron Decision Trees for Imbalanced Drifting Data Streams.
Bartosz KrawczykPrzemyslaw SkryjomskiPublished in: ECML/PKDD (2) (2017)
Keyphrases
- cost sensitive
- data streams
- class imbalance
- decision trees
- concept drift
- naive bayes
- cost sensitive learning
- class distribution
- decision tree algorithm
- misclassification costs
- probability estimates
- sliding window
- classification algorithm
- binary classification
- cost sensitive classification
- multi class
- fraud detection
- learning algorithm
- base classifiers
- change detection
- active learning
- minority class
- neural network
- imbalanced data
- training data
- machine learning algorithms
- data sets
- support vector
- data distribution
- classification trees
- training set
- classification accuracy
- probability estimation
- boosting algorithms
- non stationary
- decision rules
- support vector machine
- text classification
- bayesian networks
- outlier detection