Implement natural language processing solutions →mediumMultiple Choice

AI-102 Practice Question: Implement natural language processing solutions

Exhibit

Refer to the exhibit.

{
  "displayName": "MyConversationApp",
  "analysisInput": {
    "conversationItem": {
      "text": "Book a flight from Seattle to New York for tomorrow",
      "id": "1",
      "participantId": "user1"
    }
  },
  "parameters": {
    "projectName": "FlightBooking",
    "deploymentName": "production"
  },
  "kind": "Conversation"
}

You are testing a Conversational Language Understanding application. You send the JSON request shown in the exhibit. What is the purpose of this request?

⚠ Common exam trap

Many exam-takers confuse the purpose of CLU (intent/entity analysis) with generative AI or other NLP services, assuming any language input to Azure AI implies translation, summarization, or response generation, when in fact CLU is strictly a classification and extraction engine.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Analyze the utterance for intent and entities.

The JSON request sends a user utterance to a Conversational Language Understanding (CLU) endpoint, which is designed to analyze natural language input. The response will include the predicted intent (e.g., 'GetWeather') and extracted entities (e.g., 'location: Seattle'), fulfilling the core function of CLU: intent and entity recognition. This is not a generative or translation task; it is a classification and extraction operation.

Answer analysis

Option-by-option breakdown

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

✗
Translate the text to another language.
Why it's wrong here
Conversational Language Understanding maps utterances to intents and entities within the trained language; it performs no language translation. Translation is handled by Azure AI Translator, which would be the right choice if the request needed text rendered into a different target language.
✗
Generate a response to the user.
Why it's wrong here
Conversational Language Understanding returns predicted intents and extracted entities; it does not synthesise replies. Response generation belongs to a question-answering or orchestration layer, such as Azure AI Language's custom question answering or an LLM, which would be the right pick if the stem asked for a chatbot reply.
✗
Summarize the conversation.
Why it's wrong here
Conversational Language Understanding performs intent classification and entity extraction on utterances; it holds no conversation history and produces no summaries. Summarisation is a separate Azure AI Language feature, conversation summarisation, which would be correct if the stem described condensing a transcript.
✓
Analyze the utterance for intent and entities.
Why this is correct
The prediction request submits an utterance to the Conversational Language Understanding runtime, returning the top-scoring intent plus any extracted entities. It performs inference only; authoring, training and deployment occur through separate project and model endpoints.

Go deeper

AI-102 Practice Question: Implement natural language processing solutions

Exhibit

Why each option matters

Option-by-option breakdown

Related to this question