AI-103 Practice Question 49: Opinion Mining Aspect Sentiment
Microsoft Azure AI Apps and Agents Developer Associate (AI-103) · Test 5 · question 9 of 10
A hotel chain analyzes guest reviews with Azure AI Language sentiment analysis. One review reads: "The room was spotless and quiet, but the front-desk staff were rude and check-in took forever." Overall sentiment analysis returns "mixed" for the review, but the operations team needs to know specifically that the complaint targets staff and check-in, not the room, so it can route the review to the right department.
What should the team add to the request?
Show answer and explanation
Answer:
Opinion mining, also called aspect-based sentiment analysis, is "a feature of sentiment analysis" that "provides more granular information about the opinions related to words (such as the attributes of products or services) in text." Turning it on returns each target the review mentions ("front-desk staff," "check-in") paired with its own sentiment and the assessment (opinion) word attached to it, such as "rude" or "forever," which is exactly the per-aspect signal the operations team needs to route the complaint correctly.
- B is wrong because key phrase extraction surfaces salient noun phrases but attaches no sentiment to any of them; it can't tell the team which phrase was praised and which was criticized. - C is wrong because a more precise confidence score still describes sentiment at the document or sentence level as a whole; it doesn't decompose "room" from "staff" when both appear in the same sentence. - D is wrong because named entity recognition labels spans by entity type (person, organization, and so on) but carries no sentiment polarity, so it can't tell the team that the staff mention was negative and the room mention was positive.