# Google Knowledge Graph
**Entity class:** Search knowledge representation system
## Definition and history
Google's Knowledge Graph joined structured entity information to search experiences and named Freebase as one of its sources. [[articles/Intelligence Inside|Intelligence Inside]] uses the Freebase lineage to describe how language interfaces came to refer to machine-representable objects and relations. A query about an entity can draw on linked properties rather than only matching words in documents. The graph is one institutional implementation of structured representation, not the whole field of knowledge graphs.
## Mechanism and limits
Entity disambiguation determines which real-world object a phrase denotes, while relations constrain what can be answered about it. Errors of identity or edge attribution can propagate into answers even when sentence generation is fluent. Preserving provenance across ingestion and migration is therefore part of the interface's reliability.
## Relationships
**Input lineage:** [[wiki/Freebase|Freebase]]. **Related representation:** [[wiki/UMBEL|UMBEL]]. **Interface domain:** search and question answering. **Article route:** [[articles/Intelligence Inside|Intelligence Inside]].
## Sources
[Google Research, *From Freebase to Wikidata*](https://research.google/pubs/from-freebase-to-wikidata-the-great-migration/); [[articles/Intelligence Inside|Intelligence Inside]].