An efficient parameter estimation method for generalized Dirichlet priors in naïve Bayesian classifiers with multinomial models.
Tzu-Tsung WongChao-Rui LiuPublished in: Pattern Recognit. (2016)
Keyphrases
- parameter estimation
- bayesian classifiers
- dirichlet distribution
- em algorithm
- bayesian classifier
- bayes classifiers
- logistic regression
- expectation maximization
- maximum likelihood
- bayes classifier
- model selection
- least squares
- mixture model
- probabilistic model
- prior knowledge
- supervised classification
- naive bayes
- parameter learning
- markov random field
- maximum likelihood estimation
- machine learning
- bayesian network classifiers
- generative model
- bayesian framework
- naive bayes classifier
- similarity measure
- computer vision
- posterior distribution
- parametric models
- prior information
- text categorization
- maximum a posteriori
- decision trees
- cost sensitive learning
- image processing
- data mining