Minimum classification error rate training of supervised topic mixture model for multi-label text categorization.
Zhiyang HePing LvJi WuPublished in: ISCSLP (2014)
Keyphrases
- text categorization
- multi label
- mixture model
- unsupervised learning
- feature selection
- feature selections
- multi label classification
- supervised learning
- text classification
- gaussian mixture model
- em algorithm
- text classifiers
- generative model
- gaussian mixture
- semi supervised learning
- probabilistic model
- knn
- model selection
- k nearest neighbor
- image annotation
- language model
- semi supervised
- expectation maximization
- naive bayes
- training set
- maximum likelihood
- unlabeled data
- feature set
- text documents
- latent dirichlet allocation
- learning algorithm
- topic models
- class labels
- probability density function
- markov random field
- neural network
- graph cuts
- support vector