Question

A logistics company wants to build a lightweight information-extraction app that pulls the shipment ID, carrier name, and delivery date from thousands of scanned bill-of-lading documents. Before running any documents through Azure Content Understanding in Foundry Tools, what must the team define so the service knows which pieces of information to return?

Accepted Answer

A field schema (analyzer) that specifies the named fields to extract from each document

Answer

A prompt template with a system message that describes the company's tone of voice

Answer

A fine-tuning dataset of thousands of labeled example completions for a chat model

Answer

An embedding index that stores vector representations of every document page