AI-103: Developing AI Apps and Agents on AzureDemo

By Microsoft · Question Mode

QuestionQ31

Implement information extraction solutions

You have an application that processes scanned PDF invoices. The invoices use varied layouts and include tables that span multiple pages.

Your pipeline uses optical character recognition (OCR) to extract totals and invoice numbers, but the results are frequently incorrect because the document's structure is ignored.

You need a solution that provides OCR, layout analysis, and field extraction that generalizes across templates. The solution must NOT require training a custom model, and it must minimize administrative effort.

What should you include in the solution?

  • A Azure Language in Foundry Tools
  • B Azure Content Understanding in Foundry Tools
  • C an Azure Machine Learning model
Explanation

For high-variation, semi-structured documents such as invoices from multiple vendors with inconsistent layouts, Microsoft recommends the Azure Content Understanding custom analyzer: it combines OCR, layout analysis, and field extraction, generalizes well across template and language variation, and requires no labeled training data to get started, which minimizes administrative effort. Azure Language in Foundry Tools performs text analytics rather than layout-aware field extraction, and building a solution on an Azure Machine Learning model would require training and maintaining a custom model, which the requirement explicitly rules out.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!