Maximum entropy and least square error minimizing procedures for estimating missing conditional probabilities in Bayesian networks.
Parag C. PendharkarPublished in: Comput. Stat. Data Anal. (2008)
Keyphrases
- maximum entropy
- conditional probabilities
- bayesian networks
- probabilistic inference
- joint distribution
- incomplete data
- probabilistic model
- probability distribution
- maximum entropy principle
- random variables
- probabilistic logic
- bayesian network structure
- conditional random fields
- probabilistic knowledge
- belief networks
- missing data
- posterior probability
- transformation based learning
- structure learning
- bayesian belief networks
- markov networks
- graphical models
- naive bayes
- machine learning
- missing values
- credal networks
- knn
- approximate inference
- probabilistic reasoning
- generative model
- information extraction