Efficient integration of generative topic models into discriminative classifiers using robust probabilistic kernels.
Koffi Eddy IhouNizar BouguilaWassim BouachirPublished in: Pattern Anal. Appl. (2021)
Keyphrases
- topic models
- generative model
- discriminative classifiers
- discriminative learning
- probabilistic model
- probabilistic topic models
- latent dirichlet allocation
- text documents
- generative process
- bayesian networks
- parameter learning
- latent variables
- highly efficient
- text mining
- co occurrence
- multiple kernel learning
- object categories
- efficient learning
- bayesian network classifiers
- kernel function
- posterior probability
- information retrieval
- em algorithm
- graphical models
- object detection
- semi supervised
- multi class
- knowledge discovery
- hidden markov models
- pairwise
- feature space
- feature selection