SAFRAN (Scalable and fast non-redundant rule application) is a framework for fast inference of groundings and aggregation of predictions of logical rules in the context of knowledge graph completion/link prediction. It uses rules learned by AnyBURL (Anytime Bottom Up Rule Learning), a highly-efficient approach for learning logical rules from knowledge graphs.

Paper preprint on arXiv • AKBC 2021 conference paper (for citations)

Documentation

Can be found here.

Citation

@inproceedings{
  ott2021safran,
  title={{SAFRAN}: An interpretable, rule-based link prediction method outperforming embedding models},
  author={Simon Ott and Christian Meilicke and Matthias Samwald},
  booktitle={3rd Conference on Automated Knowledge Base Construction},
  year={2021},
  url={https://openreview.net/forum?id=jCt9S_3w_S9},
  doi={}
}

This project received funding from netidee.