Topic Model Kernel: An Empirical Study towards Probabilistically Reduced Features for Classification.
Tien-Vu NguyenDinh Q. PhungSvetha VenkateshPublished in: ICONIP (2) (2013)
Keyphrases
- topic models
- feature space
- feature vectors
- feature set
- feature extraction
- topic modeling
- latent dirichlet allocation
- support vector
- svm classifier
- class labels
- kernel function
- co occurrence
- feature selection
- machine learning
- generative model
- text mining
- text documents
- latent variables
- pattern classification
- baseline models
- supervised learning
- support vector machine
- learning algorithm
- image classification
- prior knowledge
- training set
- word pairs