Filter- versus wrapper-based feature selection for credit scoring.
Petr SomolBart BaesensPavel PudilJan VanthienenPublished in: Int. J. Intell. Syst. (2005)
Keyphrases
- credit scoring
- feature selection
- wrapper method
- logistic regression
- support vector machine svm
- support vector machine
- support vector
- sequential forward
- classification accuracy
- credit card
- text categorization
- feature ranking
- classification and regression trees
- data mining technology
- machine learning
- information entropy
- text classification
- data mining
- knn
- k nearest neighbor
- decision trees
- feature space
- multi class
- feature subset
- naive bayes
- unsupervised learning
- factor analysis
- feature set
- feature extraction
- risk assessment
- dimensionality reduction
- nearest neighbor
- social network analysis