When you process documents in test mode, the documents are moved to the validation queue regardless of whether fields need validation or data extraction was successful so that developers can analyze the data extraction results.

Prerequisites

Procedure

  1. Log in to your Control Room.
  2. Navigate to AI > Document Automation.
  3. Choose one of the following options to validate documents:
    • The View status option: Click the View status option for the learning instance that has test mode enabled or click the learning instance, click view status, and click a request that you want to validate.
    • The Validate documents option: Click the Validate documents option and click a task that you want to validate.

      A number next to the Validate documents link indicates that the learning instance has documents waiting for validation.

    Important: Documents processed in test mode will always be in the validation queue because they are used to compare extraction results. See Process documents in test mode.
    The Automation Co-Pilot validator user interface is displayed. See Validating documents in test mode using Automation Co-Pilot validator.
  4. Optional: You can validate a different version of the document by selecting a version from the version history drop-down.
    If the version you selected does not have a document to validate, you can select the Process now option to process the document using the selected version.
  5. Check each field to verify the data type and extracted value.
    Document Automation supports these data types: text, number, date, address, and check box.

    Alternatively, from the drop-down list on the right panel, you can select Show fields that need validation.

    Note: Data extracted for fields using generative AI providers is indicated using the Icon to represent the data was extracted using generative AI provider icon.
  6. Fix the fields with errors.
    Click the field or draw a box around the values that you want to extract.
    For Automation Anywhere pre-trained models, you can set up the learning instance to extract specific values in a field and ignore others. For more information, see Use validation feedback to extract specific values in a table.
  7. After making the necessary corrections, click Submit & Reprocess to verify the extracted data.
    After you validate the results, you can close the tab.