What's changed: Added per-section figures (cert-figure-retrofit). New AI-103 Chapter 6 (Azure Content Understanding = multimodal structured extraction/analyzer & schema/prebuilt vs custom/build flow, generative-prompt extraction of entities/topics/summaries = structured output/grounding & evaluation/structured vs schema extraction)
6.1Information Extraction with Azure Content Understanding
Understand implementing Azure Content Understanding to extract structured fields/insights from documents, forms, images, audio, and video—defining analyzers and schemas (extraction fields), prebuilt vs custom, and the build flow—from a developer’s view.
Azure Content Understanding extracts needed information in structured form from diverse content—documents, forms, images, audio, video (also covered in AI-901). At AI-103’s implementation level you create an analyzer (the whole extraction setup), define a schema (extraction field names/types), analyze files, and receive structured results (e.g., JSON). Use prebuilt for generic extraction and custom to define your own fields for org-specific forms.
| Step | What you do |
|---|---|
| 1. Create analyzer | For the target content type |
| 2. Define schema | Specify fields (name, type) |
| 3. Analyze | Pass files and run analysis |
| 4. Get results | Structured data with fields filled |
Common: (1) "structured extraction of fields from invoices/forms/images/audio/video" = Azure Content Understanding. (2) "define extraction fields" = the schema (part of the analyzer). (3) generic = prebuilt, org-specific forms = custom. (4) flow = analyzer → schema → analyze → structured results. (5) distinguish OCR (text recognition, Chapter 4) from structured extraction (typed fields).
Continue reading — free sign-up
You're reading the free preview. Sign up free to read this section in full, plus every chapter (including 4+) and all questions.