Knowledge graph
A knowledge graph is a structured representation of knowledge in graph form. Entities such as people, places, organizations, concepts, or events are represented as nodes, and the relationships between them as labeled edges. A knowledge graph aims to make facts and relationships explicit in a form that both people and computer systems can process.
On this page
Source content checked . English translation published 2026-10-27.
In business automation
A knowledge graph represents entities and their relationships, not just a list of documents. In automation, it helps when you need to know that a client is related to a contract, a supplier, and an open inquiry, rather than simply find a similar paragraph.
Examples
These examples illustrate a possible workflow. Adapt the connections and rules to your own systems.
Linking a client to a contract and inquiry
An agent is asked about a renewal. The system retrieves the relationship between the client, the active contract, and an open service inquiry. The answer gives the contract number from the graph instead of inventing a connection based on a general paragraph in a guide.
Relevant tools
- LlamaIndex
- You can build structured retrieval and work with documents; a full knowledge graph requires entity modeling, not just splitting text into vectors.
What to check
- A graph that is not updated after a contract change can return an old relationship as if it were still correct.