A new algorithm for data discretization and feature selection.
Marcela Xavier RibeiroAgma J. M. TrainaCaetano Traina Jr.Published in: SAC (2008)
Keyphrases
- computational complexity
- data sets
- noisy data
- input data
- feature selection
- detection algorithm
- irrelevant features
- database
- information loss
- np hard
- preprocessing
- data analysis
- learning algorithm
- optimal solution
- neural network
- data reduction
- missing data
- dynamic programming
- probabilistic model
- classification accuracy
- synthetic datasets
- worst case
- feature set
- training data
- expectation maximization
- feature subset
- objective function
- data structure
- data sources
- feature weighting
- k means
- extracted features
- machine learning
- spectral clustering
- synthetic data
- similarity measure
- model selection
- text classification
- knowledge discovery
- data points
- probability distribution