evidoria

← Back to browse

Good practice Imported

data.norge.no AI Search — Norway's Natural-Language Discovery Layer for the National Data Catalogue

Norway · Oslo · See the Norway profile

Norway's national open-data catalogue, data.norge.no (run by Digdir), added an AI-powered natural-language search built on Google Vertex AI, letting users ask questions like 'Which areas in Norway are protected?' instead of guessing exact dataset titles or technical terms.

Details

Maturity
Established
Promoter
Digdir (Norwegian Digitalisation Agency) — data.norge.no / Felles datakatalog
Period
ongoing
Keywords
open data, natural-language search, government digitalisation

Context

data.norge.no (Felles datakatalog) is Norway's official catalogue of datasets, APIs, concepts and information models contributed by public and private organisations, maintained by Digdir, the Norwegian Digitalisation Agency.

Objectives

The portal's AI search feature lets users query the catalogue's metadata in natural language rather than needing to know exact dataset names, technical terms or formats.

Activities

Digdir's own documentation explains that the service stores all search queries and hit counts — not linked to individuals — to analyse and improve search over time, and explicitly advises users not to enter personal data in their queries.

Results

Digdir's documentation itself cautions that the AI search 'may be both incomplete and contain errors,' and no public accuracy, adoption or before/after discoverability figures have been published.

Conclusions

The feature is live in production across the whole national catalogue rather than a limited pilot, but its measured impact on data discoverability has not been demonstrated.

Implementation
Indicative cost
Low (< €50k) — No budget figures disclosed; described as a documented feature added to an existing platform run by a national digitalisation agency.
Time to results
Short (< 1 year) — Feature is live and ongoing in production; no specific launch date given beyond 'ongoing.'
Staffing & skills
Digdir (Norwegian Digitalisation Agency) — platform owner and operator

Conditions for success

  • An existing, well-populated national data catalogue to search over
  • Clear user guidance not to input personal data into search queries
  • Logging of queries and hit counts (anonymised) to enable ongoing improvement

Common failure modes

  • No published accuracy or adoption metrics exist to validate real-world impact
  • Digdir's own documentation warns results may be incomplete or contain errors

Where it fits

Governance type
national digitalisation agency
Scale
national
Income level
high-income

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful