Density estimation-based document categorization using von Mises-Fisher kernels.
Andrew SkabarShahmeer MemonPublished in: IJCNN (2010)
Keyphrases
- density estimation
- document categorization
- mixture model
- text categorization
- text classification
- probability density function
- kernel function
- kernel methods
- gaussian mixture model
- text documents
- generative model
- em algorithm
- unsupervised learning
- document clustering
- model selection
- probabilistic model
- machine learning
- outlier detection
- feature space
- feature selection
- information retrieval
- expectation maximization
- support vector