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.

• ✗

  Key Phrase Extraction

  Why it's wrong here

  Key Phrase Extraction pulls out the most relevant phrases in a document based on context and frequency, helping marketers quickly see the main topics, such as 'delivery speed' or 'customer service.' However, these extracted phrases are not assigned any sentiment score or label—the service doesn't know if a phrase is associated with praise or complaint. While key phrases complement sentiment analysis by revealing themes, they no more measure emotional tone than a table of contents reveals a book's theme.

  When this WOULD be correct

  A question asks: 'Which Azure AI Language feature should be used to automatically extract the most important topics or concepts from a large collection of customer feedback forms?'

• ✓

  Sentiment Analysis

  Why this is correct

  Sentiment Analysis is the correct choice because Azure AI Language's sentiment analysis capability is explicitly built to assess the emotional tone of text and label it positive, negative, or neutral. The marketing team can feed thousands of product reviews, comments, or survey responses into this service and receive aggregated sentiment scores at document and sentence levels. This directly answers the requirement to 'automatically analyze' customer opinions for emotional polarity, making it the right tool.

• ✗

  Named Entity Recognition

  Why it's wrong here

  Named Entity Recognition (NER) scans text to locate and classify named entities such as people, organizations, locations, dates, and products, and can enrich analysis with entity links. In a marketing scenario, NER might identify which brand names appear in reviews, but it produces zero data about the polarity—whether those references are positive, negative, or neutral. Since the team's aim is to gauge subjective opinion at scale, NER cannot satisfy the sentiment detection task despite being useful for entity-level reporting.

  When this WOULD be correct

  A question asks: 'Which Azure AI Language feature should be used to extract names of products, companies, and locations from customer feedback?'

• ✗

  Language Detection

  Why it's wrong here

  Language Detection determines which natural language a text is written in, returning a language code and confidence score, but it does not assess the emotional tone or valence of that content. For example, knowing a review is 'en' tells the marketing team the text is English, yet provides no insight into whether the reviewer is praising or criticizing a product. While useful to route documents for translation or further NLP, this service fails the core goal of extracting sentiment from customer feedback.

  When this WOULD be correct

  A company receives customer feedback in multiple languages and needs to route each message to the appropriate language-specific support team. 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?”

Go deeper

Related to this question