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.

• ✗

  Sentiment Analysis

  Why it's wrong here

  Sentiment Analysis is wrong because it assigns a sentiment score and label (positive, negative, neutral, or mixed) to each sentence or document, based on how emotional or opinionated the language is. It does not summarize what the text is about; a negative score might flag an angry customer but not whether the complaint concerns latency, a dead battery, or a refund delay. The support team would still need a separate step to extract the actual words and phrases describing the technical problem, which is exactly what Key Phrase Extraction does.

  When this WOULD be correct

  A question asking to automatically determine whether customer feedback is positive, negative, or neutral (e.g., 'Analyze product reviews to gauge overall customer satisfaction').

• ✓

  Key Phrase Extraction

  Why this is correct

  Key Phrase Extraction is the appropriate Azure AI Language feature for this task because it uses a statistical model to scan the entire transcript and return the most salient, topic-bearing phrases — e.g., 'slow internet connection' or 'billing error' — across all customer utterances. This directly surfaces the underlying issues the support team needs to route, prioritize, or act on. Unlike the other options, it does not require predefined categories or sentiment labels; it is designed precisely to distill descriptive problem statements from unstructured text.

• ✗

  Named Entity Recognition (NER)

  Why it's wrong here

  Named Entity Recognition (NER) is not suitable because NER is trained to identify and classify rigid, predefined semantic categories — persons, organizations, locations, dates, quantities, and other proper-noun-like entities. Customer issues are typically expressed as descriptive multi-word phrases (e.g., 'account locked after password change') that do not fit these fixed entity types. NER would return 'Jasmine' as a person and 'billing' as a skill, but it would fail to group 'payment not going through' into a meaningful problem statement, whereas Key Phrase Extraction is designed to return exactly that.

  When this WOULD be correct

  A question asks: 'Which Azure AI Language feature should be used to extract specific entity types such as person names, dates, and organizations from customer emails?' In that case, NER would be correct.

• ✗

  Language Detection

  Why it's wrong here

  Language Detection is incorrect because it only identifies the dominant natural language of each transcript (e.g., 'English' or 'Spanish') using character- and word-level patterns. It performs a simple classification over a handful of language codes and gives no information about the content, topics, or problems customers discuss. While language metadata might help route transcripts to the right local-language model, it cannot reveal 'internet is slow' or 'payment failed' as actionable issues.

  When this WOULD be correct

  A multinational company receives customer feedback in multiple languages and needs to route each message to the appropriate language-specific support team. Language Detection would be the correct feature to automatically determine 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