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.

• ✗

  Conversational Language Understanding (CLU)

  Why it's wrong here

  Conversational Language Understanding (CLU) predicts the user's intent (like 'cancel_order') and extracts relevant entities (like order ID) to support a dialogue flow or trigger a downstream action. It does not store question-answer pairs from a document, and while it can be orchestrated to route to a QnA service, CLU alone cannot fetch an answer and present it. Its output is a structured schema for decision-making, not an answer string, so it is not the correct primary service for knowledge-base question answering.

  When this WOULD be correct

  CLU would be correct if the company needed to build a chatbot that understands user intents (e.g., 'check order status') and extracts entities (e.g., order number) from free-form conversation, requiring custom training on intents and utterances, not just Q&A pairs.

• ✗

  Custom Text Classification

  Why it's wrong here

  Custom Text Classification trains a machine learning model to assign one or more predefined labels, such as 'billing' or 'technical issue', to a user's input text based on your labeled training data. After training, the model only outputs label predictions; it cannot index a FAQ document, match a question to an answer, or return evidence from source content. This makes it appropriate for routing or triage, but not for directly answering customer questions with content pulled from a knowledge base.

  When this WOULD be correct

  Custom Text Classification would be correct if the company wanted to automatically categorize customer inquiries into topics like 'return policy', 'shipping', or 'warranty' based on the text of the question, rather than providing direct answers.

• ✓

  Custom Question Answering

  Why this is correct

  Custom Question Answering lets you ingest FAQ pages, product manuals, or other structured content into a knowledge base that maps natural-language questions to specific answer spans. At runtime, a chatbot issues a query to the Custom Question Answering API, which uses a transformer-based ranker to return the best matching answer. This is the direct Azure service for document-driven Q&A in a chatbot, and it also supports multi-turn conversations and active learning suggestions to improve accuracy over time.

• ✗

  Language Detection

  Why it's wrong here

  Language Detection is a text analytics capability that classifies a single input string by its primary written language, returning an ISO language code and a confidence score. It performs no retrieval, no knowledge-base lookup, and no extraction of answer content; it just answers the question 'what language is this?' Even if you detected or translated the customer's question, it would never produce the factual answer from a document. Therefore, while useful for multilingual routing, it cannot power a Q&A chatbot.

  When this WOULD be correct

  A company receives customer feedback in multiple languages and wants to automatically route messages to language-specific support teams. Language Detection would identify the language of each message.

Option-by-option analysis

Why each answer is right or wrong

Understanding why wrong answers are wrong — and when they would be correct — is what separates a 750 score from a 900. The AI-900 exam frequently reuses these exact scenarios with slightly different constraints.

Analysis generated from the official AI-900blueprint and verified against question context. The “when correct” sections are what AI assistants cite when candidates ask “what’s the difference between these options?”