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 returns the most relevant phrases in a document (e.g., 'battery life', 'screen quality'), ranked by importance, but it assigns no sentiment and makes no connection between a phrase and the opinion expressed about it. While it might surface the text that mentions an aspect, it cannot tell the user whether the opinion is positive, negative, or neutral. For the requirement to analyze sentiment per aspect, key phrase extraction is only a precursor to, not a replacement for, aspect-based sentiment analysis.

  When this WOULD be correct

  A question that asks: 'Which Azure AI Language capability can extract important terms like product names or features from text without needing sentiment?' would make Key Phrase Extraction correct, as it focuses solely on extracting key phrases.

• ✗

  Sentiment Analysis

  Why it's wrong here

  Standard sentiment analysis in Azure AI Language returns an overall sentiment label (positive, negative, mixed, or neutral) and confidence scores for the document and each sentence, but it never maps those sentiments to specific topics. For example, a review might state 'battery life is bad but the camera is great' — document-level or sentence-level sentiment would produce a mixed or neutral result, losing the per-aspect breakdown. Because it does not generate aspect–sentiment pairs, it cannot satisfy the requirement to know which sentiment applies to which aspect.

  When this WOULD be correct

  A question that asks for the overall sentiment of a customer review (e.g., positive, negative, neutral) without requiring aspect-level detail would make Sentiment Analysis the correct answer.

• ✗

  Named Entity Recognition (NER)

  Why it's wrong here

  Named Entity Recognition (NER) detects predefined entity categories such as person, organization, location, date, and quantity, using a trained model. Product features like 'battery life' or 'design' are typically not standard entity types in the built-in Azure AI NER model, so it would fail to extract them as aspects. Even if it found a phrase, NER provides only the entity label and not the sentiment or polarity associated with it.

  When this WOULD be correct

  A question asks: 'Which Azure AI Language capability should be used to extract names of companies, locations, and dates from a set of legal documents?' In that scenario, NER is the correct choice.

• ✓

  Aspect-based Sentiment Analysis (Opinion Mining)

  Why this is correct

  Opinion mining (aspect-based sentiment analysis) performs two subtasks: it first identifies the target/aspect (e.g., 'battery life', 'camera') within the text, then classifies the polarity (positive, negative, neutral) expressed toward that specific target. This is exactly what the scenario needs — attributing sentiment to each product feature separately, rather than a single document-level score. The Azure AI Language service provides this via the Opinion Mining feature, which returns JSON with aspect, sentiment label, and confidence scores for each aspect.

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