Multi-labeled document classification using semi-supervived mixture model of Watson distributions on document manifold.
Nguyen Kim AnhNgo Van LinhNguyen Khac ToiNguyen The TarnPublished in: SoCPaR (2013)
Keyphrases
- document classification
- mixture model
- probabilistic model
- gaussian mixture model
- gaussian densities
- von mises fisher
- text categorization
- dirichlet distribution
- em algorithm
- density estimation
- web documents
- text documents
- dirichlet process
- generative model
- classification algorithm
- unsupervised learning
- text classification
- text mining
- maximum likelihood
- language model
- automatic document classification
- expectation maximization
- gaussian mixture
- model selection
- supervised learning
- training data
- topic models
- neural network
- document clustering
- data points
- high dimensional
- object recognition
- search engine