NTB Tagger · Entity Tagging API
Precision tagging for
newsroom-grade content.
Tagger is NTB's entity recognition and metadata enrichment API. It identifies persons, locations, events and organisations in text — linked to Wikidata's knowledge graph and classified with IPTC Media Topics codes.
Sample output
Erna SolbergPER addressed parliament in OsloLOC regarding NATO Summit 2025EVT, on behalf of HøyreORG.
IPTC: 11000000 · Politics · 20000530 · International relations
Capabilities
Everything a newsroom needs.
Identify and extract named individuals from any text. Linked to Wikidata profiles for authoritative metadata.
Detect geographic entities — cities, countries, regions — and enrich them with geocoordinates and hierarchical data.
Recognise named events such as elections, summits, and disasters. Ideal for contextual indexing and archive search.
Extract institutions, companies, and bodies. Cross-referenced with Wikidata for legal names and classification.
Every entity is resolved against Wikidata's knowledge graph, returning structured identifiers and rich linked data.
Automatically classify content with IPTC Media Topics codes — the global standard for news content metadata.
Integration
One API call. Structured metadata.
Submit text or article
Send a POST request to the Tagger API with your article body, headline or any freeform text.
Entities are extracted
Our NLP pipeline detects and classifies entities across all supported types in milliseconds.
Receive structured metadata
Get a clean JSON response with tags, confidence scores, IPTC codes and Wikidata identifiers ready to ingest.
POST /v1/tag
{
"text": "Erna Solberg addressed parliament in Oslo regarding NATO Summit 2025.",
"lang": "no",
"types": ["PER", "LOC", "EVT", "ORG"],
"enrich": ["wikidata", "iptc"]
}Get started
Ready to enrich your content?
Tagger is available to NTB partners and media organisations. Request access and start tagging in minutes.