Bitzu

AI-901 ยท Information extraction with Content Understanding

What is an analyzer in Azure Content Understanding in Foundry Tools?

What Azure Content Understanding in Foundry Tools does, the idea of an analyzer with a field schema, prebuilt vs custom analyzers, and the structured output it produces.

Key points

  • An analyzer is the configuration that defines how one kind of content is processed: how content is extracted, which field schema to fill, and which models to use.
  • Feed it a document, image, audio or video file and it returns structured output (Markdown and/or JSON that matches your schema).
  • Prebuilt analyzers cover common cases; custom analyzers add your own fields.

Source: https://learn.microsoft.com/en-us/azure/ai-services/content-understanding/overview

Practice questions

Single choice

Which input types can Azure Content Understanding in Foundry Tools turn into structured output?

  1. Only scanned documents and PDF forms
  2. Documents, images, audio and video
  3. Only text typed into a chat prompt
  4. Only database tables and CSV files
Show answer

Answer: B. Documents, images, audio and video

It handles four modalities: documents, images, audio and video. The other options name only a subset or unrelated sources.

Select all that apply

Which statements about the output of Content Understanding are correct?

  1. It can produce Markdown suited to search and RAG scenarios
  2. It can produce JSON that matches the field schema you define
  3. It can attach a confidence score between 0 and 1 to extracted values
  4. It only ever returns one plain-text transcript
  5. You must write a long prompt for every single field
Show answer

Answer: A. It can produce Markdown suited to search and RAG scenarios, B. It can produce JSON that matches the field schema you define, C. It can attach a confidence score between 0 and 1 to extracted values

Output can be Markdown for search and retrieval or schema-shaped JSON for automation, and fields carry a 0-1 confidence when enabled. It is not limited to plain text and does not require prompt engineering for field extraction.

True or false

True or false: prebuilt analyzers are fixed and cannot be used as a starting point for your own analyzer.

Show answer

Answer: False

False. You can fetch a prebuilt analyzer's definition, edit it and save it as your own custom analyzer.

Practice 8 questions on this topic in Bitzu

Short lessons, instant explanations and spaced review of what you get wrong. Free on iOS and Android.

Download on the App StoreGet it on Google Play

More AI-901 topics in this module