Combining multi-distributed mixture models and bayesian networks for semi-supervised learning.
Manuel StrittLars Schmidt-ThiemeGerhard PoeppelPublished in: ICMLA (2007)
Keyphrases
- semi supervised learning
- mixture model
- unsupervised learning
- bayesian networks
- semi supervised
- probabilistic model
- gaussian mixture model
- unlabeled data
- generative model
- labeled data
- supervised learning
- em algorithm
- unlabeled samples
- expectation maximization
- semi supervised classification
- density estimation
- co training
- machine learning
- maximum likelihood
- active learning
- training data
- model selection
- language model
- graphical models
- cluster analysis
- incomplete data
- labeled and unlabeled data
- label propagation
- dimensionality reduction
- transfer learning
- pairwise constraints