Introduction
The Helmholtz Knowledge Graph (Helmholtz KG) is a central tool developed within the Helmholtz Metadata Collaboration (HMC) to support measurable improvements in metadata quality, interoperability, and visibility across Helmholtz research infrastructures.
The KG aggregates metadata from diverse Helmholtz data repositories, libraries and publication systems, and research information systems into a coherent, linked, and queryable representation. It serves as a unified semantic foundation for understanding how data, software, publications, institutions, experts, and research projects relate to one another. Further the KG itself stands as an independent asset. It is both a diagnostic tool (observing metadata quality and consistency across the association) and a web service that supports scientific discovery and cross-infrastructure integration.
Curious about How Your Metadata Is Processed?
If the above made you curious about how your metadata is harvested or mapped, this documentation provides insights into the underlying processes. The KG infrastructure processes metadata in a deliberately neutral and transparent way, assembling the graph based on what is provided by source systems and how those inputs align with shared schemas and mappings. Thus, it "unhides" the status of metadata with the Helmholtz association - therefore we operate all our services under *unhide.helmholtz-metadaten.de- domains.
Observed issues in the KG — such as missing links, inconsistencies, duplication, or unexpected representations — are often the result of interactions between metadata creation at the source and processing across pipeline stages. Understanding these interactions is essential for improving metadata quality sustainably.
We see the Helmholtz-KG as a starting point for dialogue rather than a final verdict. Whether you operate a data service, steward metadata, or consume KG data, we are happy to discuss observations and explain pipeline behavior.
Have questions about how your metadata appears in the Helmholtz Knowledge Graph or want to better understand the processing pipelines?
📧 hmc@fz-juelich.de
You can also open an issue or start a discussion in our GitLab project to report observations or suggest improvements:
🔗 https://codebase.helmholtz.cloud/hmc/hmc-public/unhide/development/feedback
