AI-102 Question 314
Single answerYou are working on a customer support chatbot that uses Azure Cognitive Services QnA Maker for its knowledge base. After creating and populating the knowledge base, you need to make it available for production use. Which step must you take to publish the knowledge base?
- A
Export the knowledge base to a JSON file and upload it to the Azure Web App hosting the bot.
- B
Use the 'Publish' option in the QnA Maker portal to make the knowledge base accessible via an endpoint.
- C
Integrate the knowledge base with Azure Bot Service before it is automatically published.
- D
Configure API Management to expose the knowledge base through a secure API.
Show answer and explanation
Correct answer: B
Explanation
To make a QnA Maker knowledge base available for production, you must use the 'Publish' option in the QnA Maker portal. This action creates a REST API endpoint that can be used by applications, such as bots, to query the knowledge base. Other options, such as exporting to JSON or using API Management, are unrelated to the publishing process.
- A. Incorrect.
Exporting the knowledge base to a JSON file is not a valid method for publishing it. Publishing in QnA Maker involves using the 'Publish' option in the portal, which creates an accessible endpoint.
- B. Correct.
Using the 'Publish' option in the QnA Maker portal is the correct step to make the knowledge base accessible via an endpoint. This is required for production use.
- C. Incorrect.
Integrating with Azure Bot Service is a separate task that can be done after the knowledge base is published. It is not part of the publishing process itself.
- D. Incorrect.
API Management can be used to secure and manage APIs, but it is not required or used specifically for publishing a QnA Maker knowledge base.