AZ-900 Describe Azure architecture and services Practice Question

Answer analysis

Option-by-option breakdown

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

• ✗

  Azure Text Analytics

  Why it's wrong here

  Azure Text Analytics is a set of natural language processing features that perform analyses on raw text: sentiment scores, key-phrase extraction, named entity recognition (like people, places, dates), and language detection. It neither learns a custom intent model from user utterances nor extracts user-defined entities. For instance, it can tell you the sentiment of 'I want to reset my password' is neutral and highlight 'password' as a recognized subject, but it does not determine that the user's goal is to reset the password. That goal classification is exactly LUIS's job.

• ✓

  Azure Language Understanding (LUIS)

  Why this is correct

  Azure Language Understanding (LUIS) is purpose-built to interpret natural language in interactive applications: you define intents (like 'OrderCoffee' or 'GetWeather') and entities (like 'size' or 'location'), then train the model with labeled example utterances. It learns to recognize variations, synonyms, and implicit context, and returns a JSON payload with the top-scoring intent and extracted entities for downstream code to act on. That is precisely the type of intelligent understanding needed to drive a dialog bot that can fulfill user requests.

• ✗

  Azure Translator

  Why it's wrong here

  Azure Translator is a machine translation service that converts an entire sentence from one language to another, such as English-to-French, while preserving intended meaning as closely as possible. It operates solely at the language surface level — it changes the form, not the interpretation of the utterance. Even after translating 'Book a flight to Seattle,' the system still doesn't know that the user wants an action called 'BookFlight' with an entity 'Seattle'. LUIS, in contrast, abstracts the entire sentence into a structured semantic representation independent of the words used.

• ✗

  Azure QnA Maker

  Why it's wrong here

  Azure QnA Maker (now QnA in Azure AI Language) is designed to extract question-and-answer pairs from FAQ documents, PDFs, and spreadsheets, and then return the matched answer when a user asks a similar question. It does not parse an utterance to classify an overall 'intent' or extract custom 'entities' that the application can act upon. In a conversational app, QnA Maker answers 'how do I reset my password?' but fails to trigger an orchestrated action beyond retrieving a static answer. LUIS would instead classify that same text as 'FindResetGuide' intent and extract 'password' as the target.

Go deeper

Related to this question