A generative model for unsupervised discovery of relations and argument classes from clinical texts.
Bryan RinkSanda M. HarabagiuPublished in: EMNLP (2011)
Keyphrases
- generative model
- probabilistic model
- bayesian framework
- em algorithm
- prior knowledge
- mixture model
- discriminative models
- semi supervised
- posterior probability
- latent dirichlet allocation
- discriminative learning
- rhetorical structure theory
- topic models
- object categories
- fisher kernel
- latent variables
- hierarchical bayesian model
- generative and discriminative models
- video representation
- image segmentation
- object recognition
- multiscale