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.

• ✗

  Document-level is more accurate; opinion mining is a faster but less precise approximation

  Why it's wrong here

  This is a false dichotomy: document-level sentiment and opinion mining are not different algorithms competing on accuracy or speed. Both are produced by the same underlying model in Azure AI Language, and opinion mining simply performs additional word-level analysis to associate sentiments with specific aspects of the text. Neither is an approximation of the other—document-level returns one overall label, while opinion mining returns per-aspect labels. You might choose one over the other based on the level of insight you need, but you are not trading accuracy for speed.

• ✓

  Document-level assigns one overall sentiment; opinion mining extracts per-aspect sentiments

  Why this is correct

  Document-level sentiment analysis in Azure AI Language evaluates an entire text and returns a single sentiment label (positive, negative, mixed, or neutral) with confidence scores. Opinion mining, however, goes further by running aspect-based analysis that identifies specific targets in the text—such as 'food' or 'service'—and assigns each its own sentiment. This makes opinion mining far more actionable: for a mixed review like 'The food was great but the service was slow,' document-level gives a mixed score, while opinion mining reveals food: positive and service: negative. The core distinction is granularity, not capability.

• ✗

  Opinion mining works only on social media posts; document-level works on all text types

  Why it's wrong here

  The claim that opinion mining only works on social media posts is incorrect. Both document-level sentiment analysis and opinion mining are powered by the same pre-built Azure AI Language API and accept any raw text, including product reviews, support tickets, survey responses, and forum posts. The actual difference lies in output granularity: document-level condenses the whole text into one sentiment score, whereas opinion mining extracts sentiment per mentioned aspect. Neither tool is constrained by source type or content channel.

• ✗

  Document-level sentiment requires training data; opinion mining is pre-built

  Why it's wrong here

  This option misrepresents how Azure AI Language works. Both document-level sentiment analysis and opinion mining are pre-built capabilities of the Sentiment Analysis API—they require no training data, no custom models, and no labeled examples from the developer. You simply call the API and choose the 'opinion mining' flag to get aspect-level results. Document-level sentiment is not a custom ML solution; it is a standard, out-of-the-box feature that works immediately on your text. The only thing that differs is the level of detail returned, not the need for training.

Go deeper

Related to this question