Feature selection for regression problems based on the Morisita estimator of intrinsic dimension.
Jean GolayMichael LeuenbergerMikhail F. KanevskiPublished in: Pattern Recognit. (2017)
Keyphrases
- regression problems
- intrinsic dimension
- feature selection
- high dimensional data
- dimension reduction
- linear regression
- dimensionality reduction
- multi task
- cross validation
- least squares
- genetic programming
- intrinsic dimensionality
- nearest neighbor
- maximum likelihood
- model selection
- support vector machine
- hyperparameters
- input space
- high dimensional
- data sets
- support vector
- text categorization
- feature extraction
- low dimensional
- similarity search
- machine learning
- feature space
- sparse representation
- data points
- maximum a posteriori
- data analysis
- unsupervised learning
- data distribution
- linear discriminant analysis
- feature subset
- text classification
- pattern recognition
- random forests
- learning machines
- feature set
- knn
- input data
- multi class
- principal component analysis