Nonconvex Regularizations for Feature Selection in Ranking With Sparse SVM.
Léa LaporteRémi FlamaryStéphane CanuSébastien DéjeanJosiane MothePublished in: IEEE Trans. Neural Networks Learn. Syst. (2014)
Keyphrases
- feature selection
- feature ranking
- support vector
- high dimension
- support vector machine
- ranking svm
- sparse pca
- low rank matrices
- ranking algorithm
- feature space
- knn
- group lasso
- multi class
- learning to rank
- text categorization
- ranking functions
- selected features
- text classification
- feature selection algorithms
- feature set
- optimization problems
- classification performances
- high dimensional
- machine learning
- support vector machine svm
- bayes classifier
- dimensionality reduction
- classification accuracy
- objective function
- reduced set
- robust principal component analysis
- multi task
- information gain
- convex optimization
- microarray data
- sparse representation
- feature subset
- web image annotation
- web search
- small sample
- variable selection
- text classifiers
- input features
- regression model
- gene selection
- kernel function
- decision trees
- multiple kernel learning
- feature extraction
- training data
- signal recovery
- feature vectors
- minimization problems
- classification method
- model selection
- classification algorithm
- svm classifier
- fold cross validation