New feature selection method for multi-class data: Iteratively weighted AUC (IWA).
Petr HonzíkPavel KuceraOndrej HyncicaDaniel HauptPublished in: IDAACS (1) (2011)
Keyphrases
- multi class
- feature selection
- support vector machine
- data sets
- pairwise
- test data
- classification accuracy
- missing data
- training samples
- feature subset
- missing values
- multi class classifier
- error correcting output codes
- cross validation
- logistic regression
- data points
- high dimensional
- learning algorithm
- model selection
- text classification
- feature vectors
- training data
- multi task
- multi class classification
- binary classifiers
- single class
- similarity measure
- multi class svm