Implement an Azure AI Search Solution AI-102 Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   An e-commerce site wants to offer 'similar products' search. They have product descriptions and want to find items with similar meaning. They plan to use Azure AI Search. What should they do?

   • AEnable semantic search and use the built-in re-ranker.
   • BUse a wildcard query on the product description field.
   • CCreate an index with a vector field, generate embeddings for product descriptions, and query using vector search.
   • DAdd a filter on the product category field.
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A real estate company indexes property listings. They want to automatically extract the number of bedrooms from the description text. No built-in skill provides this. They have an HTTP endpoint that returns the bedroom count. What should they do?

   • AUse the Entity Recognition skill to extract the bedroom count.
   • BUse the Key Phrase Extraction skill to identify bedroom numbers.
   • CAdd the HTTP endpoint as a data source and index its responses.
   • DCreate a custom skill that calls the HTTP endpoint, add it to a skillset, and map its output to a field in the index.
   Reveal Solution Discussion Select an answer first
4. 13foundation · easy

   What is required to use vector search in an Azure AI Search index?

   • AFields of type Collection(Edm.Single) that store vector embeddings
   • BA data source of type Azure Cosmos DB
   • CA custom skill to generate vectors
   • DA semantic search configuration
   Reveal Solution Discussion Select an answer first
5. 14expert · hard

   A team is building a skillset that includes a custom skill. The custom skill sometimes fails due to timeouts. They want to ensure that indexing continues even if the custom skill fails, but they also want to know which documents failed. What should they do?

   • ASet the indexer to fail on the first error and restart it manually.
   • BConfigure the custom skill to allow failures and enable indexer logging to track failed documents.
   • CRemove the custom skill from the skillset to avoid failures.
   • DIncrease the timeout of the indexer to allow the custom skill to complete.
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A company has a SQL database with customer records. They create an Azure AI Search index and a data source. They want the index to be populated with the current data. What should they do next?

   • ACreate an indexer that references the data source and the index, then run it.
   • BCreate a custom skill that reads from SQL and writes to the index.
   • CUse the Azure portal to manually upload the SQL data to the index.
   • DEnable semantic search to automatically populate the index.
   Reveal Solution Discussion Select an answer first