Question 36
Domain 2: Implement Knowledge Mining and Azure AI Search SolutionsA law firm has 500,000 PDF contracts stored in Azure Blob Storage. They need to: extract text, identify entities (parties, dates, amounts), and make the content searchable. The solution must be built without custom code. Which Azure AI Search components should you configure?
Correct answer: B
Explanation
Azure AI Search uses a “data source” to connect to Blob Storage, an “indexer” to pull in the PDFs, and an “index” to store searchable fields. A “skillset with built-in skills” provides built-in enrichment such as text extraction and entity recognition, so the content can be made searchable “without custom code.”
Why each option is right or wrong
A. Index + Indexer (no skillset needed)
B. Index + Data Source + Indexer + Skillset with built-in skills
Azure AI Search requires a data source to point at Azure Blob Storage, an indexer to crawl the 500,000 PDFs and populate the search pipeline, and an index to hold the searchable fields; these are the core objects defined in the Azure AI Search indexing model. To extract text and recognize entities such as people, organizations, dates, and amounts without custom code, you attach a skillset using built-in cognitive skills—specifically document cracking/text extraction plus entity recognition—so the enrichment happens during indexing rather than in an external app.
C. Index + Custom skill (Azure Function) only
D. Document Intelligence + Azure AI Search (separate pipelines)