Feature Selection for High-Dimensional Data: A Kolmogorov-Smirnov Correlation-Based Filter.
Jacek BiesiadaWlodzislaw DuchPublished in: CORES (2005)
Keyphrases
- high dimensional data
- kolmogorov smirnov
- feature selection
- dimensionality reduction
- high dimensionality
- preprocessing step
- dimension reduction
- nearest neighbor
- high dimensional
- low dimensional
- data sets
- data points
- subspace clustering
- data analysis
- data distribution
- similarity search
- high dimensional datasets
- goodness of fit
- text categorization
- user satisfaction
- feature extraction
- knn
- feature space
- linear discriminant analysis
- missing values
- mutual information
- support vector machine
- clustering high dimensional data
- machine learning
- high dimensional spaces
- text classification
- support vector
- pattern recognition
- principal component analysis
- test statistic
- computer vision
- neural network
- information gain
- euclidean distance
- model selection
- input data
- semi supervised
- upper bound
- classification accuracy
- k means