A company needs to extract entities from a PDF document to build a classifier model. Which solution will extract and store the entities in the LEAST amount of time?

Question

Options

• AUse Amazon Comprehend to extract the entities. Store the output in Amazon S3.
• BUse an open source AI optical character recognition (OCR) tool on Amazon SageMaker to extract
• CUse Amazon Textract to extract the entities. Use Amazon Comprehend to convert the entities to
• DUse Amazon Textract integrated with Amazon Augmented AI (Amazon A2I) to extract the entities.

• A
  22% (11)
• B
  8% (4)
• C
  60% (30)
• D
  10% (5)

Explanation

Option C combines Amazon Textract (for PDF text extraction via built-in OCR) with Amazon Comprehend (for entity detection), giving you a fully managed, serverless pipeline with no infrastructure setup - the fastest path from PDF to labeled entities.

Why the distractors fail:

• A is incomplete - Comprehend requires plain text input and cannot read PDFs directly, so you'd still need an OCR step first.
• B uses open-source OCR on SageMaker, which requires provisioning compute, installing packages, and writing custom code - far more setup time than managed AWS services.
• D adds Amazon A2I, which introduces a human review loop for low-confidence predictions; this is great for accuracy but explicitly adds time, contradicting the "least amount of time" requirement.

Memory tip: Think of the pipeline as two lanes - Textract sees the document (OCR), and Comprehend understands it (NLP/entities). When you need speed, always reach for fully managed services over custom SageMaker setups, and avoid A2I whenever the question asks to minimize time.

No community discussion yet for this question.