A trustworthy natural-language interface over any ontology-described data space
Plain-language questions become deterministic, ontology-compliant SPARQL — and the only thing that has to change to support a new domain is the ontology, not the code.
Press → or Space to navigate · 1) purpose · 2) architecture & standards · 3) the ontology-artifact core
Purpose · 30 seconds
Data spaces like ENVITED-X already publish deeply structured asset metadata as ontologies (OWL) and constraints (SHACL). That richness is wasted if reaching it requires SPARQL, prefixes, and schema expertise.
Governed, standards-based metadata: classes, shapes, allowed values, cross-references — already curated for interoperability.
Users think in "German motorways with 3 lanes", not in sh:targetClass, prefixes, and hand-assembled graph patterns.
Search must stay explainable, safe, and reproducible — convenience cannot come at the cost of correctness.