What it means
A knowledge graph represents information as a network of entities, such as people, places, organisations and objects, and the relationships between them. It makes connections explicit and machine-readable, rather than leaving them hidden in text.
Real-world example
German libraries describe authors, organisations and subjects using the Gemeinsame Normdatei (GND), an authority file that works as a large network of linked entities. A museum links its records to the same identifiers, so a search for a particular photographer connects her works across the museum, partner libraries and archives, even where her name is spelled differently.
Why it matters for AI adoption
Knowledge graphs provide structure and reliable relationships that language models alone do not. Combining them with AI can improve precision, consistency and explainability, especially in heritage, research and administrative data. Building and maintaining a good graph requires sustained curation.
Questions to ask
- Do we already use authority files or controlled vocabularies we could build on?
- Who would maintain the entities and relationships over time?