Data Flow for Standard Forms in Document Automation

The following data flow diagram shows how to configure Standard Forms in Document Automation using various components that facilitate the flow of data between Document Automation and the third-party extraction service.


The flow of data between Document Automation and Standard forms
  • Labeling and model creation tool

    You use the model creation tool to label and create models. In this diagram, the training documents are uploaded in the tool and stored in the blob storage of the third-party service. To obtain the best results, we recommend that you upload and label a minimum of five training documents to create a model.

  • Blob storage

    Blob storage is a storage unit that contains training documents and its related information.

  • Extraction model

    The trained models stored in the blob storage of the third-party service are fetched through APIs. The Model ID, which can be a GUI ID or a standard name, is used to create the learning instance.

  • Learning instance creation

    You can browse and select the model to create a Learning instance. After you select the model, you can upload a document for extraction to this Learning instance. The Learning instance accesses the third-party extraction service to retrieve extraction results. The results either move to the output folder, or the document moves to the validator.

  • Extraction engine

    The extraction engine performs extraction based on the model ID and returns the results to the Learning instance.

  • Output results
    Once the extraction is completed, the output results will include three components:
    • extraction .csv
    • auto detected table .csv
    • response.json