Parametrizations and reference priors for multinomial decomposable graphical models.
Guido ConsonniHélène MassamPublished in: J. Multivar. Anal. (2012)
Keyphrases
- graphical models
- probabilistic model
- markov networks
- exponential family
- belief propagation
- probabilistic inference
- random variables
- probabilistic graphical models
- conditional random fields
- structure learning
- approximate inference
- bayesian networks
- map inference
- conditional independence
- factor graphs
- expectation maximization
- text classification
- bayesian framework
- belief networks
- prior knowledge
- graph structure
- exact inference
- text categorization
- mixture model
- message passing
- feature selection
- topic models
- statistical relational learning
- upper bound
- em algorithm
- generative model
- latent variables
- prior information
- naive bayes