AI-102 Question 344
Select 3You are designing an Azure AI Search solution for a large e-commerce platform. The platform has millions of products, and users should be able to search for products using natural language queries, such as 'Show me laptops under $1000 with 16GB RAM'. The search results should include facets for filtering (e.g., price range, brand) and support for synonyms (e.g., 'notebook' for 'laptop'). Which combination of steps is required to implement this solution?
- A
Create an Azure Cognitive Search service and configure an index to store the product data.
- B
Enable semantic search in the Azure Cognitive Search service to support natural language queries.
- C
Define a synonym map in the Azure Cognitive Search service to handle synonyms like 'notebook' and 'laptop'.
- D
Configure Azure Blob Storage as the data source and enable automatic indexing for real-time updates.
- E
Use Azure Machine Learning to train a custom natural language processing (NLP) model for the search.
Show answer and explanation
Correct answers: A, B, C
Explanation
To implement the Azure AI Search solution, you need to set up an Azure Cognitive Search service with a well-configured index to store and query product data. Enabling semantic search allows the search service to process natural language queries effectively, and defining a synonym map ensures users can find results even when using different terms for the same concept. Azure Blob Storage or a custom NLP model is not required for this scenario, as the built-in capabilities of Azure Cognitive Search are sufficient.
- A. Correct.
Correct: An Azure Cognitive Search service and a properly configured index are essential for storing and querying the product data.
- B. Correct.
Correct: Enabling semantic search allows the service to understand user intent and process natural language queries.
- C. Correct.
Correct: Synonym maps are specifically designed to handle query terms that have equivalent meanings, like 'notebook' and 'laptop'.
- D. Incorrect.
Incorrect: While Azure Blob Storage can be used as a data source, automatic indexing is not a requirement for this scenario, especially when the data source is not real-time.
- E. Incorrect.
Incorrect: Azure Cognitive Search already has built-in capabilities for natural language queries and semantic search, so training a custom NLP model is unnecessary.