Saber (Sequence Annotator for Biomedical Entities and Relations) is a deep-learning based tool for information extraction in the biomedical domain.

The neural network model used is a BiLSTM-CRF [1, 2]; a state-of-the-art architecture for sequence labelling. The model is implemented using Keras / Tensorflow.

The goal is that Saber will eventually perform all the important steps in text-mining of biomedical literature:

  • Coreference resolution (✅)
  • Biomedical named entity recognition (BioNER) (✅)
  • Entity linking / grounding / normalization (✅)
  • Simple relation extraction (🔜)
  • Event extraction (🔜)

Pull requests are welcome! If you encounter any bugs, please open an issue in the GitHub repository.