Maximum entropy and maximum likelihood criteria for feature selection from multivariate data.
Sankar BasuCharles A. MicchelliPeter OlsenPublished in: ISCAS (2000)
Keyphrases
- maximum entropy
- multivariate data
- maximum likelihood
- feature selection
- high dimensional
- maximum entropy principle
- categorical data
- principal components
- random fields
- likelihood function
- em algorithm
- high dimensionality
- parameter estimation
- feature space
- conditional random fields
- parameter space
- original data
- minimum cross entropy
- principle of maximum entropy
- feature extraction
- dimensionality reduction
- machine learning
- transformation based learning
- support vector machine
- feature set
- expectation maximization
- text classification
- dimension reduction
- multi task
- multivariate time series
- support vector
- high dimensional data
- least squares
- model selection
- pattern recognition
- generative model
- cluster analysis
- unsupervised learning