AI-102 Question 293
Select 3You are designing a custom question answering solution using Azure AI Language for a company’s internal knowledge base. The knowledge base contains product manuals and troubleshooting guides. The company wants employees to query this knowledge base in natural language and receive accurate answers. Which steps should you include in your implementation?
- A
Import the product manuals and troubleshooting guides into an Azure Blob Storage container, then configure Azure Cognitive Search to index the data.
- B
Use the Azure AI Language Studio to create a custom question answering project and upload the knowledge base documents.
- C
Train the model using labeled intent and entity data to improve the accuracy of responses.
- D
Enable the question answering feature in Azure AI Language and test the solution with sample queries.
- E
Deploy the solution as an Azure Bot Service to allow employees to interact with the knowledge base via a chatbot.
Show answer and explanation
Correct answers: B, D, E
Explanation
To create a custom question answering solution using Azure AI Language, you need to upload the knowledge base documents into the Azure AI Language Studio and enable the question answering feature. Testing the solution ensures that it performs as expected. Finally, deploying the solution as an Azure Bot Service allows employees to interact with it in a user-friendly manner.
- A. Incorrect.
Importing the documents into Azure Blob Storage and using Azure Cognitive Search can be part of indexing documents for general search functionality but is not specific to building a custom question answering solution in Azure AI Language.
- B. Correct.
This step is essential for creating a custom question answering solution as it involves uploading and preparing the knowledge base documents in the Azure AI Language Studio.
- C. Incorrect.
Training the model with labeled intent and entity data is not required for the custom question answering feature, as it does not rely on intent or entity recognition but instead focuses on extracting answers from knowledge base documents.
- D. Correct.
Enabling the question answering feature and testing with sample queries is a critical step in validating that the solution works as expected.
- E. Correct.
Deploying the solution as an Azure Bot Service is necessary to provide an interactive interface for employees to query the knowledge base in natural language.