Human-in-the-Loop Overview

Human-in-the-Loop (HITL) enables human verification and corrections to ensure accuracy of data extracted by Document AI processors before it is used in critical business applications.

It provides a workflow and UI for humans (referred to as labelers in HITL) to review, validate and correct the data extracted from documents by Document AI processors. It is used across Financial Services, Health, Manufacturing, Government and other industries.

We provide two options of human labelers:

  • Bring-your-own-labelers - your own workforce or partner workforce can review the documents.
  • Google HITL Workforce - is available as a Private Preview service for documents. Fill out this form to request access

Features

  • Confidence threshold filters to limit the number of documents going through HITL.
  • Labeler pool management, including task assignments and efficiency analytics by task and by labeler.
  • UI cues and features that reduce labeler handling time per document.
  • Analytics and metrics by task and by labeler, so you can streamline HITL operations.

Benefits

  • Risk mitigation - mitigate financial risks of critical data being incorrect - for example, invoice amounts, billing addresses, loan amounts, etc.
  • Simplify Exception Handling - Easily roll out a human review and exception handling workflow.
  • Workforce Efficiencies - manage, monitor and improve productivity of workforce managing human review.
  • Cost control - control costs of human review with configurable filters.
  • Data completeness - ensure extracted data is complete for downstream business applications.

Processors supported

General Processors

Specialized Processors

Custom Processors

You will see the HITL configuration menu for these processors. Refer to the Release notes for announcements about additional processor support.