Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Relationships allow organizations to model, manage, and govern semantic connections between business entities. Clearly defined relationships help organizations turn complex connections into actionable insights and decisions with the following benefits:
- Semantic clarity: Explicitly defined relationships (such as owns, located at, supplies, or monitored by) allow organizations to represent not only what entities exist, but how they interact.
- Analytics: Ontologies that are enriched with relationships offer contextualized insights.
Important
This feature is in preview.
Prerequisites
Before adding relationship types to your ontology, make sure you have the following prerequisites:
- A Fabric workspace with a Microsoft Fabric-enabled capacity.
- Ontology item (preview) enabled on your Fabric tenant.
- An ontology (preview) item with entity types created.
- Relationship source data that meets these guidelines:
- The data is in OneLake.
- The source data contains keys for both the source and target entity type.
Key concepts
Relationship types use the following ontology (preview) concepts. For definitions of these terms, see the Ontology (preview) glossary.
- Entity type
- Relationship type
- Relationship instance
Create relationship type
The first step in adding a relationship in your ontology (preview) item is creating a relationship type. Then, bind data to the relationship type to create relationship instances.
For example, suppose you want to define a relationship between the entity types Truck and Driver, and your data contains a table called Truck data with columns TruckId, Site, TruckName, and DriverId. You might start by defining a relationship type called drives from the Driver entity type to the Truck entity type. Then, create a data binding based on the Truck data table, using the columns TruckId and DriverId to define relationship instances for that relationship type. The result is that your drives relationship type has instances to represent each combination of TruckId and DriverId in your data.
Follow these steps to create a relationship type and bind data to it:
Select Add relationship in the menu ribbon. Or, highlight an entity type in the Entity Types pane and select ..., then Add relationship.
In the Add relationship type to ontology window,
- Enter a Relationship type name.
- Select the Source entity type and Target entity type for your relationship. The source and target entity types must be distinct from one another.
Select Add relationship type.
The Relationship configuration pane opens. In this pane, you define the columns from the source data that connect instances of these entity types.
Under Source data, select your workspace, lakehouse, and table that contains the keys for both your target and source entity type.
For each entity type, select a Source column from the linking source data that identifies instances of that entity type. The source column selections must match the entity type keys.
Tip
If you don't see any keys for an entity type, make sure your source and target entity types have keys defined.
Select Create.
Verify the new relationship type is visible on the configuration canvas.
Edit or delete relationship type
You can edit or delete relationship types that exist in your ontology (preview) items in the Relationship configuration pane.
To edit a relationship type, select it in the configuration canvas. Then, update any of the fields in the Relationship configuration pane.
To delete a relationship type, select it in the configuration canvas. Then, select the Delete relationship type button in the Relationship configuration pane.
Note
Any updates in upstream data sources (like new rows) need to be manually refreshed before they're visible in the ontology item. For more information, see refresh the graph model.