Feature selection based on maximizing separability in Gauss mixture model and its application to image classification.
Sangho YoonRobert M. GrayPublished in: ICIP (2) (2005)
Keyphrases
- mixture model
- image classification
- feature selection
- em algorithm
- feature extraction
- model selection
- unsupervised learning
- gaussian mixture model
- dimensionality reduction
- generative model
- density estimation
- expectation maximization
- text categorization
- bag of words
- probabilistic model
- maximum likelihood
- mutual information
- feature set
- language model
- feature space
- feature selection algorithms
- pattern recognition
- mixture modeling
- image features
- machine learning
- image representation
- text classification
- classification accuracy
- support vector
- visual words
- probability density function
- sparse coding
- dirichlet process
- object recognition
- reinforcement learning
- feature subset
- clustering algorithm