BioVAE: a pre-trained latent variable language model for biomedical text mining.
Hai-Long TrieuMakoto MiwaSophia AnaniadouPublished in: Bioinform. (2022)
Keyphrases
- language model
- latent variables
- pre trained
- biomedical text mining
- probabilistic model
- text mining
- language modeling
- training data
- graphical models
- n gram
- speech recognition
- topic models
- training examples
- query expansion
- information retrieval
- generative model
- semi automated
- test collection
- mixture model
- expectation maximization
- control signals
- prior knowledge
- bayesian networks
- smoothing methods
- random variables
- text classification
- data mining
- machine learning
- word sense disambiguation
- statistical model
- natural language processing
- dimensionality reduction
- data sets
- em algorithm