Implement an Azure AI Search Solution AI-102 Practice Questions (Page 6)

Questions 26–30

1. 26application · medium

   A medical research team indexes journal articles. They need to extract patient age from the text, but no built-in skill provides this. They have an Azure Function that accepts text and returns age. How should they integrate this into the enrichment pipeline?

   • ACreate a custom skill that calls the Azure Function, then add the custom skill to a skillset and reference it in an indexer.
   • BCall the Azure Function directly from the search query to compute age at query time.
   • CAdd the Azure Function as a data source and use an indexer to pull its output.
   • DUse the Azure Function to pre-process documents before they are uploaded to the data source.
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   Which types of projections can be defined in an Azure AI Search knowledge store?

   • ADocument, image, and video projections
   • BIndex, skillset, and indexer projections
   • CFile, object, and table projections
   • DBlob, queue, and table projections
   Reveal Solution Discussion Select an answer first
4. 28foundation · easy

   Which action would you take to load data from an Azure SQL database into an existing Azure AI Search index?

   • ACreate a new index with the same schema
   • BModify the skillset to include a SQL connector
   • CRun the indexer that is configured with the data source and index
   • DSend a search query to the index
   Reveal Solution Discussion Select an answer first
5. 29application · medium

   A company stores product documentation as PDFs in an Azure Storage account. They need to make the text searchable through an Azure AI Search index. The storage account is in a different region from the search service. You must minimize the amount of code and infrastructure to set up. What should you do?

   • ACreate an indexer that connects directly to the storage account using a shared access signature, but do not create a data source.
   • BUse Azure Data Factory to copy the PDFs into the search service, then create an index manually.
   • CWrite a custom application that reads each PDF, extracts text, and pushes documents to the index using the Azure SDK.
   • DCreate an index, then use the Azure portal to import data by selecting the storage account as the data source and enabling the indexer.
   Reveal Solution Discussion Select an answer first
6. 30expert · hard

   A company has an Azure AI Search index populated from a SQL database. They update the database nightly. They want the index to reflect the changes automatically. They also need to ensure that deleted rows are removed from the index. What should they do?

   • AConfigure the indexer to run on a schedule and enable soft-delete detection on the data source.
   • BRun the indexer manually every night after the database update.
   • CUse a custom skill to detect changes and update the index.
   • DEnable semantic search to automatically sync the index with the database.
   Reveal Solution Discussion Select an answer first