A scalable implementation of information theoretic feature selection for high dimensional data.
Anthony KleerekoperMichael PappasAdam Craig PocockGavin BrownMikel LujánPublished in: IEEE BigData (2015)
Keyphrases
- information theoretic
- high dimensional data
- mutual information
- feature selection
- high dimensionality
- dimensionality reduction
- information theory
- dimension reduction
- high dimensional
- nearest neighbor
- low dimensional
- information bottleneck
- theoretic framework
- jensen shannon divergence
- subspace clustering
- data points
- data sets
- manifold learning
- linear discriminant analysis
- data analysis
- similarity search
- data distribution
- low rank
- input space
- information theoretic measures
- clustering high dimensional data
- high dimensional spaces
- sparse representation
- feature extraction
- knn
- distributional clustering
- missing values
- support vector machine
- variable selection
- text categorization
- text classification
- high dimensional datasets
- input data
- image registration
- pattern recognition
- support vector
- neural network