AI-102 Question 361
Select 3You are developing an Azure Cognitive Search solution for a retail application. The application stores product data in an Azure SQL Database. You need to create an index to enable full-text search on the product catalog where customers can search by product name, category, and description. Which steps should you take to create and configure the index?
- A
Define the index schema, specifying fields such as product name, category, and description with appropriate attributes like searchable and filterable.
- B
Use the Azure CLI command 'az search index create' to create the index after defining the index schema in JSON format.
- C
Set up a skillset in Azure Cognitive Search to preprocess the product data and map the data to the index fields.
- D
Configure an indexer to connect to the Azure SQL Database and populate the index with product data.
- E
Manually populate the index by uploading data through the Azure portal.
Show answer and explanation
Correct answers: A, B, D
Explanation
To successfully create an index in Azure Cognitive Search, you need to define the index schema, create the index using tools like the Azure CLI, and configure an indexer to populate the index with data from the source. Skillsets are not relevant when working with structured data, and manual data uploads are not recommended for production environments.
- A. Correct.
This is correct because the index schema defines the structure of the index and specifies which fields are searchable, filterable, or retrievable.
- B. Correct.
This is correct because Azure CLI can be used to create an index by providing the schema in JSON format.
- C. Incorrect.
This is incorrect because a skillset is used for cognitive enrichment tasks, such as extracting information from unstructured data, but is not necessary for creating or configuring an index for structured SQL data.
- D. Correct.
This is correct because an indexer automates the process of pulling data from a data source (such as Azure SQL Database) and populating the index.
- E. Incorrect.
This is incorrect because manually uploading data is not a scalable approach, especially for large datasets, and does not align with best practices in Azure Cognitive Search.