AI-102 Question 294
Select 3You are designing a custom question answering project in Azure AI Language to provide tailored responses to customer inquiries about your company's product catalog. Which steps must you complete to successfully create and publish the custom question answering project?
- A
Import your existing FAQ documents or manually add question-and-answer pairs.
- B
Define intents and entities to structure the knowledge base.
- C
Train the custom question answering model after finalizing the knowledge base content.
- D
Publish the project to make it available for querying.
- E
Enable Azure OpenAI Service to process natural language queries.
Show answer and explanation
Correct answers: A, C, D
Explanation
To create and publish a custom question answering project, you must first build the knowledge base by adding question-and-answer pairs, train the model to process the knowledge base, and finally publish the project to make it available for use. Steps like defining intents/entities or enabling Azure OpenAI Service are not applicable to this specific Azure AI feature.
- A. Correct.
Correct. You need to provide question-and-answer pairs, either by importing existing FAQ documents or manually entering them, to build the knowledge base for the custom question answering project.
- B. Incorrect.
Incorrect. Defining intents and entities is a step required for Azure Bot Service or Language Understanding (LUIS) but is not part of creating a custom question answering project.
- C. Correct.
Correct. Training the model ensures that the custom question answering system understands the provided knowledge base and can respond accurately to queries.
- D. Correct.
Correct. Publishing the project is necessary to make the custom question answering model accessible for client applications or end-users to query.
- E. Incorrect.
Incorrect. Enabling Azure OpenAI Service is not required for a custom question answering project, as it uses Azure AI Language service instead.