Cost-Sensitive back-propagation neural networks with binarization techniques in addressing multi-class problems and non-competent classifiers.
Zhongliang ZhangXinggang LuoSalvador GarcíaFrancisco HerreraPublished in: Appl. Soft Comput. (2017)
Keyphrases
- multi class problems
- cost sensitive
- cost sensitive learning
- misclassification costs
- multi class
- naive bayes
- class imbalance
- active learning
- binary classification
- class distribution
- binary classification problems
- support vector machine
- boosting algorithms
- decision trees
- multiclass classification
- binary classifiers
- multi class classification
- training examples
- text classification
- classification accuracy
- feature selection
- training set
- linear classifiers
- logistic regression
- probability estimation
- pairwise
- feature space
- bayesian networks
- roc curve
- classification error
- classification algorithm
- data points
- machine learning