Catalog model
Groups → Packages → Resources
Organize datasets by team and project, with clear ownership, policies, and searchable metadata at the resource level.
SemanticMatter THiNK Platform
THiNKSemantic documentation and interoperability
THiNK helps you document data with shared semantics, connect it across systems, and keep the context that makes it reusable. It combines a federated catalog, ontology-based modeling, and assisted documentation in one web interface.
Governance
Provenance and access control
Track where data came from and how it was produced, and apply role-based access where it matters.
Technical stack
RDF/OWL, DC, SKOS, SHACL, PROV-O, DCAT, SKOS, FNO
Open standards you can keep using over time—inside your own stack and across partners.
Where interoperability breaks down
Inconsistent documentation
Datasets often arrive without the context people need to reuse them or trust the results.
Semantics live in people
Units, concepts, and relationships are known by the team, but not captured in a shared, formal model.
Paper still holds the evidence
Lab sheets, notebooks, and handwritten forms contain key details that never make it into digital systems.
What THiNK does
Built for teams that need strong semantics without turning every project into a knowledge engineering effort.
Shared meaning across teams
Define domain concepts with ontologies and controlled vocabularies so different groups describe data the same way.
Faster metadata capture
Turn text and tabular snippets into structured metadata, with review steps that keep the result defensible.
Human-reviewed automation
Let the system propose shapes, units, and links—then have experts confirm and adjust before anything is published.
Knowledge graph foundation
Store ontologies, mappings, and metadata in a versioned graph model that supports SPARQL and reasoning when needed.
Governance that matches ownership
Manage access and policies at the resource level, while keeping data ownership with the teams that run the sources.
FAIR in day-to-day work
Make data easier to find, access, combine, and reuse by capturing provenance and semantics as part of normal workflows.
From paper to structured metadata
Capture context from scanned notebooks, photographed forms, and handwritten notes—then attach it to the datasets it belongs to.
OCR text captureSemantic extractionOntology mappingSHACL validationGraph update
Standards and FAIR alignment
Use open standards so your metadata stays portable, machine-readable, and usable across tools and organizations.
RDF/OWLSHACLPROV-ODCATSKOSFNODC
Priority use cases
Reconstruct lab context
Turn handwritten and mixed-language lab notes into consistent metadata with traceable provenance.
Support FAIR data management
Back FAIR plans with searchable metadata, access controls, ontology alignment, and reuse-ready provenance.
Provide a backbone for digital twins
Keep models and simulations in sync with data through event-driven updates grounded in a shared knowledge graph.
Roadmap and expected impact
Lock the initial feature set
Publish user and developer docs
Run real-world pilots
Close the feedback loop
Stabilize and test end-to-end
What changes in practice
Files → Shared, queryable knowledge
Manual metadata → Assisted capture with review
Static datasets → Workflows that react to change
Paper context → Digital metadata you can search
Local reuse → Cross-team interoperability
THiNK helps you bring older, poorly documented datasets into the same workflow as newer data—without losing provenance or forcing everyone into bespoke documentation practices.