AI-102 Question 256
Select 4You are tasked with creating a language understanding model for a customer service chatbot using Azure AI Language. The chatbot needs to identify user intents such as 'Check Order Status', 'Cancel Order', and 'Update Shipping Address', and extract relevant entities like 'Order ID' and 'Address'. After creating the model, you need to ensure it can be tested and deployed for production use. What steps should you follow to correctly implement and manage this model?
- A
Define intents and entities in the Language Studio and train the model.
- B
Export the trained model and deploy it to a custom on-premises environment.
- C
Test the model using sample utterances in the Language Studio to ensure accuracy.
- D
Publish the model to an endpoint for integration with the chatbot application.
- E
Use Azure Monitor to track the model's performance after deployment.
Show answer and explanation
Correct answers: A, C, D, E
Explanation
To implement and manage a language understanding model with Azure AI Language, you need to define and train the model, test it to verify its accuracy, and publish it to an endpoint for application integration. After deployment, Azure Monitor can be used to track the model's performance and ensure it meets production requirements. Exporting the model to an on-premises environment is not a valid step in this process, as Azure AI Language is designed to run in the cloud.
- A. Correct.
Defining intents and entities in the Language Studio and training the model is a required step to create a functional language understanding model.
- B. Incorrect.
Exporting the trained model to an on-premises environment is not supported by Azure AI Language. The model is designed to be used via Azure endpoints.
- C. Correct.
Testing the model with sample utterances in Language Studio is crucial to ensure that it correctly identifies intents and extracts entities.
- D. Correct.
Publishing the model to an endpoint is necessary for integration with applications such as a chatbot.
- E. Correct.
Using Azure Monitor helps track model performance, including metrics like usage and accuracy, after deployment.