AI-900 Practice Question: Describe features of Natural Language Processing workloads on Azure

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Recognising when a user intends to cancel their subscription during a chat session

  Why it's wrong here

  Recognising when a user intends to cancel their subscription is a specific use case of intent recognition, not the definition of the capability itself. Intent recognition is the general function of classifying a user's goal from any natural language input, and cancellation is just one possible intent label among many (e.g., book, reschedule, upgrade). In an Azure AI Language conversational application, you would define a 'cancel_subscription' intent in an intents schema, but the underlying model is not limited to recognising that single action.

• ✓

  Determining the user's goal or purpose from their natural language input to route conversation logic

  Why this is correct

  Intent recognition is the Natural Language Understanding (NLU) process that extracts the user's intended objective from their utterance and maps it to a predefined intent, such as book_flight, check_weather, or get_help. The recognised intent then drives the dialog system to route the conversation to the appropriate logic or backend service. This task commonly leverages machine learning models, often built with Azure AI Language's conversational language understanding (CLU) capability, that classify the input text into intents and optionally extract entities.

• ✗

  Detecting the emotional intention behind a user's message for sentiment classification

  Why it's wrong here

  Detecting the emotional intention behind a message describes sentiment analysis, not intent recognition. Sentiment analysis evaluates the polarity of the text (positive, negative, neutral) to gauge the user's mood, whereas intent recognition identifies the user's action goal—such as booking, cancelling, checking, or finding—regardless of the emotional tone. Mixing these tasks can cause a chatbot to respond with empathy instead of taking the correct action, which is why Azure AI Language treats sentiment and intent as separate capabilities.

• ✗

  Verifying that the user's stated intent matches their historical behaviour in the application

  Why it's wrong here

  Verifying a user's stated intent against historical behaviour is a user verification, profiling, or anomaly-detection task, not intent recognition. Intent recognition operates solely on the current utterance, classifying what the user explicitly says they want to do right now. Matching past actions would require authentication and personalisation logic, not the language model itself, and is unrelated to the core definition of intent recognition.

Go deeper

Related to this question