Home/Knowledge hub/What is a knowledge graph?
GlossaryWhat is a knowledge graph?
Information is rarely useful on its own. A contract belongs to a client, a project and a person. A knowledge graph makes those links visible, for people and for AI.
In short
A knowledge graph is a network of entities, such as people, documents, clients, projects and topics, and the relationships between them. Instead of separate files in separate systems, it shows how information is connected. That helps people find the full picture faster, and it gives AI the context it needs to answer accurately. Metadata in a document management system is often the starting point.
On this page
How does a knowledge graph work?
A knowledge graph consists of entities (the nodes) and relationships (the links). Each relationship has a meaning, so the graph knows not only that two things are related, but how.
| Entity | Relationship | Related entity |
|---|---|---|
| Contract | is signed with | Client |
| Offer | belongs to | Project |
| Procedure | is owned by | Person |
| is about | Topic |
Why a knowledge graph matters for AI
Context
AI sees which documents, people and projects belong together.
Better answers
Answers reflect the full picture, not one loose document.
Fewer hallucinations
Reasoning over your own knowledge instead of generic web data.
Agents that act
AI agents follow the links to the right client, contract or owner.
Knowledge graph and metadata
In a metadata-driven document management system such as M-Files, documents are already linked to clients, projects, cases and people through their metadata. Those links are a structured basis for a knowledge graph: the relationships are recorded once and stay correct, wherever the document is shown.
Knowledge graph in Mindbreeze InSpire
By creating a connected view of enterprise knowledge, Mindbreeze InSpire identifies relationships between people, documents, systems, and topics across the organization. By linking relevant data points, teams uncover insights faster and receive answers that reflect the full picture of their enterprise knowledge.
Frequently asked questions about knowledge graphs
What is a knowledge graph?
A knowledge graph is a network of entities, such as people, documents, clients, projects and topics, and the relationships between them, so that information can be found and understood in context.
What is the difference with a database?
A classic database stores records in fixed tables. A knowledge graph puts the relationships first, so connections across systems and document types are easy to follow.
Knowledge graph or vector search?
Vector search finds text with a similar meaning. A knowledge graph knows how things are related. Modern AI solutions often combine both to find the most relevant context.
Do I have to build a knowledge graph myself?
Not necessarily. An insight engine such as Mindbreeze InSpire builds a connected view from your existing sources. Good metadata in your document management makes the result better.
Our consultants are happy to think along with you.
Plan a conversationIs your knowledge ready for AI?
We look at your sources, metadata and access rights with you, and show what is needed for reliable AI answers.
Continue learning
Related terms, articles and the solutions behind this topic.