Feature space transformations and model selection to improve the performance of classifiers.
José Ortiz-BejarEric Sadit TellezMario GraffPublished in: CoRR (2019)
Keyphrases
- model selection
- feature space
- cross validation
- meta learning
- feature selection
- hyperparameters
- regression model
- parameter estimation
- bayesian learning
- sample size
- model selection criteria
- generalization bounds
- generalization error
- feature set
- mixture model
- training set
- linear classifiers
- decision trees
- statistical inference
- variable selection
- machine learning
- selection criterion
- bayesian methods
- learning algorithm
- feature vectors
- training data
- leave one out cross validation
- error estimation
- selection criteria
- missing values
- mean shift
- machine learning algorithms
- high dimensional
- training samples
- active learning
- dimensionality reduction
- principal component analysis
- data sets
- information criterion
- kernel matrix
- feature extraction
- dimension reduction
- linear discriminant analysis
- objective function
- svm classifier
- support vector
- text classification