Learn about Automated bulk extraction and how it processes multiple documents from a configured document source manually or on a schedule.

Automated bulk extraction processes multiple documents stored in a document source, such as AWS S3, Google Cloud Storage, or Google Drive, so that you do not have to upload each document individually. A pipeline reads the documents from the source, extracts information from them by using a published extraction template, and loads the extracted data into Reltio. You can run the pipeline manually or on a schedule, and you can review the result of each run.

The following roles use Automated bulk extraction:

• Data Steward

Automated Bulk Extraction combines the following configurations:

• An extraction template defines what information to extract and how the extracted data maps to Reltio entities and attributes.
• A document source connects Unstructured augmentation to the cloud storage location that contains the files to process.
• A pipeline combines a published extraction template and a document source. It also defines the crosswalk source and an optional schedule for pipeline runs. In addition, it determines whether to create a data change request before publishing the extracted data to Reltio.

For the complete processing sequence and diagram, see Unstructured augmentation workflow.