Managing Borderline and Noisy Examples in Imbalanced Classification by Combining SMOTE with Ensemble Filtering.
José A. SáezJulián LuengoJerzy StefanowskiFrancisco HerreraPublished in: IDEAL (2014)
Keyphrases
- imbalanced data
- imbalanced data sets
- class imbalance
- binary classification problems
- imbalanced datasets
- class distribution
- minority class
- classification models
- ensemble methods
- multiple classifier systems
- class imbalanced
- training set
- training data
- classification accuracy
- decision trees
- support vector
- support vector machine
- training examples
- final classification
- feature selection
- supervised learning
- random forest
- combining classifiers
- machine learning
- text classification
- learning algorithm
- highly skewed
- neural network
- classifier ensemble
- ensemble classifier
- regression problems
- cost sensitive
- pattern classification
- class labels
- model selection
- feature set
- nearest neighbour
- decision boundary
- base learners
- data streams
- cost sensitive learning
- feature space
- ensemble learning
- machine learning methods
- classification algorithm
- benchmark datasets
- highly imbalanced