Near-Bayesian Support Vector Machines for imbalanced data classification with equal or unequal misclassification costs.
Shounak DattaSwagatam DasPublished in: Neural Networks (2015)
Keyphrases
- misclassification costs
- imbalanced data classification
- cost sensitive
- support vector
- training examples
- feature selection
- class distribution
- class imbalance
- naive bayes
- support vector machine
- binary classification
- cost sensitive learning
- active learning
- classification accuracy
- minority class
- total cost
- classification error
- multi class
- bayesian networks
- roc curve
- svm classifier
- text classification
- kernel function
- cross validation
- small number
- training set
- machine learning
- binary classifiers
- decision trees
- training data
- loss function
- multi class classification
- bayesian inference
- feature space
- support vectors
- learning algorithm
- base classifiers
- ls svm
- benchmark datasets
- test set
- multi task
- generalization ability