A law firm wants to make millions of scanned contracts searchable — extracting entities, key phrases, and making everything queryable. Which AI workload is this?
A.Anomaly detection
No outliers involved.
B.Knowledge mining
Unstructured corpus → searchable index = knowledge mining. Correct.
C.Computer vision
Vision handles image content, not full-text search scenarios.
D.Document intelligence alone
Document intelligence extracts fields from forms — a component, not the whole workload.
Why this answer
Knowledge mining = ingesting unstructured content (documents, images) and extracting searchable insights — Azure AI Search is the flagship service. Document intelligence is one INPUT capability (extracting from forms), but the end-to-end searchable-corpus scenario is knowledge mining.
Correct answer: B · Last reviewed 2026-02-14
Official AWS references
Practice this interactively
Answer without seeing the solution first in the AI Workloads & Responsible AI practice session.
Practice AI Workloads & Responsible AI