Bootstrap feature selection in support vector machines for ventricular fibrillation detection.
Felipe Alonso AtienzaJosé Luis Rojo-ÁlvarezGustavo Camps-VallsAlfredo Rosado MuñozArcadi García-AlberolaPublished in: ESANN (2006)
Keyphrases
- support vector
- feature selection
- learning machines
- cross validation
- large margin classifiers
- frequency analysis
- classification accuracy
- text categorization
- feature space
- support vector machine
- detection algorithm
- detection method
- object detection
- svm classifier
- information gain
- false alarms
- feature ranking
- logistic regression
- kernel function
- dimensionality reduction
- multi class
- loss function
- text classification
- kernel methods
- radial basis function
- feature subset
- hyperplane
- data sets
- generalization ability
- feature selection algorithms
- support vector regression
- support vector classification
- multi task
- detection rate
- false positives
- model selection
- mutual information
- training set