2.5 Build

The Build operation is the final step in the data pipeline that processes Target Entities into the knowledge graph. It transfers data from the configured Data Sources through a two-layer architecture:

1. Staging Layer – Data is first imported from the source into the staging layer, where it can be validated and prepared. Import modes include Insert (add new records), Update (modify existing records), or Preserving Merge (combine both).

2. Graph Layer – From the staging layer, data is processed into the actual knowledge graph, where nodes and Relationships become queryable.

The user can select one or more Target Entities to include in each build job, configure import parameters individually or in bulk, and optionally add related entities automatically. Once the build is complete, the resulting graph can be navigated using the Exploration module. For automated and scheduled builds, see Workflows.

Configure a Build Job

A build job processes selected Target Entities through the staging layer into the knowledge graph. Follow these steps to configure and execute a build.

Select Entities

Choose which Target Entities to include in the build job.

1. Click the Search field to look for entities for the build job.

   

2. Search for entities using a keyword.

   

3. Add the entities by selecting their checkboxes.

   

4. Continue selecting checkboxes until all required entities are selected.

   

Configure Import Parameters

Configure how each entity should be processed through the staging and graph layers.

1. Configure the import parameters for each entity. Each Target Entity has two operation categories: Staging and Graph.

   

   Staging Layer Options:

   • Import Mode: Choose how data is imported from the source into the staging layer:
     • Insert – Inserts new data into the staging layer. Existing business keys are ignored.
     • Update – Updates only existing rows with matching business keys.
     • Preserving Merge – Inserts new rows and updates existing ones without deleting unmatched rows.
   • Select Configuration – Choose the import configuration for the entity.
   • Distinct – Enables de-duplication of incoming data rows based on the business key.
   • Clear – Deletes all nodes from the staging layer before import.

   Graph Layer Options:

   • Process – Process the entity from the staging layer into the graph layer.
   • Clear – Clear the entity in the graph layer (if it already exists) before processing.

2. Select the Load Type.

   

3. Select one or multiple configurations from the dropdown.

   

Add Related Entities

Optionally include entities that are connected through relationships.

1. Use the Add related function to add related entities. Related entities (entities that are connected to this entity by a relationship) are automatically added to the build job.

   

Execute Build

Run the build job to process the configured entities.

1. After completing the build setup, confirm and run the build job by clicking Run build.

   

Bulk Configuration

When working with multiple entities, you can configure parameters for all of them at once using bulk operations.

1. Click Bulk set parameters in the top-right corner.

   

2. Select entities using checkboxes, or click Select all, to bulk set parameters.

   

3. Select the import mode: Insert / Update / Preserving Merge.

   

4. Select actions for the graph and staging layers.

   

5. Confirm the bulk edit by clicking Set.

   

6. Confirm and execute the job by clicking Run build.