A multi-view validation framework for LLM-generated knowledge graphs of chronic kidney disease
The validation framework offers a reliable and scalable method for evaluating quality and validity of LLM-generated triples across three views: semantic plausibility, type compatibility, and structural importance. The framework demonstrates robust performance in filtering high-quality triples and lays a strong foundation for fast and reliable medical KG construction and validation.
Authors
- Mario A. Cypko (ORCID: https://orcid.org/0000-0003-2944-9357)
- Oliver Amft (ORCID: https://orcid.org/0000-0001-6811-3659)
- Dilpreet Singh (ORCID: https://orcid.org/0000-0002-2176-9420)
- Aditya Kumar
Institutions
- University of Freiburg (DE)
- Hahn-Schickard-Gesellschaft für angewandte Forschung (DE)
Publication Details
- Journal
- International Journal of Computer Assisted Radiology and Surgery
- Published
- 2025-08-14
- DOI
- https://doi.org/10.1007/s11548-025-03495-x
- Primary Topic
- Biomedical Text Mining and Ontologies
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Gemeinsame Bundesausschuss