Building Heritage Digital Twins with Open-Source Tools

At EADH2026 in Kraków, we organised the practical methodology workshop Building Heritage Digital Twins with Open-Source Tools: Human-in-the-Loop Semantic Knowledge Production from Documentary Evidence.

The workshop moved from a simple question — how does documentary evidence become trustworthy, machine-actionable knowledge? — to a practical workflow for constructing Heritage Digital Twins. Using a small cultural heritage dataset, participants worked with observations, candidate semantic assertions, human review, provenance, and review states.

A central part of the workshop was hands-on experimentation with Betwixt, our open-source semantic review tool. Participants could inspect documentary evidence, review proposed classifications and relationships, correct or reject candidate claims, and experiment with multilingual descriptions. The exercises showed how the same knowledge can move between human-readable review tables and atomic semantic assertions while keeping the review process explicit.

The workshop connected these practical exercises to a broader methodology for human-in-the-loop semantic knowledge production: automated tools and AI can propose candidate knowledge, but competent human review determines what becomes part of a warranted knowledge state. Provenance, uncertainty, rights, and governance can therefore remain part of the resulting Heritage Digital Twin rather than being added afterwards.

Conference contributions

• European Association for Digital Humanities 2026. Linking Europe: Digital Humanities Without Borders. Book of Abstracts. 10.5281/zenodo.22078031.
• Presentation of our conference paper: Heritage Digital Twins Beyond Modelling. Download PDF.
• Poster: From Fragmented Heritage to Federated Digital Twins: Results from the Finno-Ugric Data Sharing Space. 10.5281/zenodo.22807041