AI-102 Question 259
Select 4You are designing a customer support chatbot using Azure AI Language. The chatbot must identify user intents such as 'CheckOrderStatus' and 'ReturnItem' and extract key entities like 'OrderID' and 'ItemName' from user input. You plan to use a Custom Language Understanding model. Which steps should you take to implement and manage this model in Azure AI Language?
- A
Create an Azure AI Language resource and configure its endpoint and key in your application.
- B
Train the model by providing labeled examples of intents and entities using the Language Studio.
- C
Deploy the model to a specific Azure region for endpoint availability.
- D
Export the model to a JSON file and upload it to the Azure Blob Storage for use.
- E
Enable autoscaling for the Azure AI Language resource to ensure optimal performance.
- F
Use the REST API to test the deployed model by sending sample user queries.
Show answer and explanation
Correct answers: A, B, C, F
Explanation
Implementing and managing a language understanding model in Azure AI Language involves creating and configuring the resource, training the model with labeled data, deploying the model for endpoint availability, and testing it using the REST API. These steps ensure the model is ready to process user queries effectively. Exporting the model to a JSON file or enabling autoscaling are not relevant actions for this scenario.
- A. Correct.
Correct: Creating an Azure AI Language resource and configuring its endpoint and key is necessary to access the model from your application.
- B. Correct.
Correct: Training the model with labeled examples of intents and entities is a required step in preparing the model to understand user input.
- C. Correct.
Correct: Deploying the model to a specific Azure region is necessary to make it available for processing requests.
- D. Incorrect.
Incorrect: Exporting the model to a JSON file and uploading it to Azure Blob Storage is not a supported step in implementing a language understanding model in Azure AI Language.
- E. Incorrect.
Incorrect: Autoscaling is not directly applicable to Azure AI Language resources, which are managed services and scale automatically based on usage.
- F. Correct.
Correct: Using the REST API to test the deployed model is a recommended practice to ensure it processes user queries correctly.