AI-102 Question 257
Select 3You are developing a conversational AI solution for a customer service chatbot. The chatbot must identify the intent behind user queries and extract specific entities, such as the 'order number' and 'issue type', using the Azure AI Language service. After creating a Language Understanding (LUIS) model, what steps should you take to publish and manage this model so it can be used by the chatbot?
- A
Train the model after labeling the data with intents and entities.
- B
Publish the model to a staging slot for testing purposes.
- C
Export the LUIS model as a JSON file and manually deploy it to the chatbot.
- D
Assign a prediction endpoint and key to the LUIS application.
- E
Directly connect the LUIS model to the bot framework without publishing it.
Show answer and explanation
Correct answers: A, B, D
Explanation
To implement and manage a language understanding model with Azure AI Language, you must train the model to learn intents and entities, publish it to a staging or production slot for usage, and configure a prediction endpoint with the necessary key for integration. These steps ensure the model is fully functional and accessible to the chatbot.
- A. Correct.
Correct: Training the model is required after labeling the data with intents and entities to ensure the model can correctly identify intents and extract entities.
- B. Correct.
Correct: Publishing the model to a staging slot allows you to test the model in a controlled environment before deploying it to production.
- C. Incorrect.
Incorrect: Exporting the LUIS model as a JSON file is not required for deployment; Azure manages this process natively.
- D. Correct.
Correct: Assigning a prediction endpoint and key is necessary for the chatbot to interact with the LUIS model and retrieve predictions.
- E. Incorrect.
Incorrect: A LUIS model must be published before it can be used by the bot framework. Direct connections without publishing are not supported.