An e-commerce company collects thousands of product reviews each month and wants to automatically compute an overall customer satisfaction trend by determining whether each review's overall tone is positive, negative, or neutral, without a human reading every review individually. Which AI capability should the team apply to each review?
Select an answer to reveal the explanation.
Short Explanation
Think about what the team actually wants to know: not what the review talks about, not what language it's in, but how the customer felt. That feeling-on-a-scale is a distinct capability from the others. Pulling out topics tells you a review mentioned battery life, but not whether the customer loved or hated the battery life. Spotting names and places tells you a brand or city was mentioned, which is useful for other purposes but says nothing about tone. Detecting the language is just housekeeping so the system knows how to process the text at all. The capability that actually reads the emotional charge of a passage and places it somewhere between glowing and furious is the one built specifically for that job, and it is the only one of the four that produces a satisfaction-style trend line when you average it across thousands of reviews.
Full Explanation
The correct answer is A. The team needs a per-review polarity score, positive, negative, or neutral, aggregated into a satisfaction trend, and that is exactly what sentiment analysis produces: it reads a passage and returns where it falls on an opinion scale. Option B is incorrect because key phrase extraction identifies the main topics or talking points in a review, such as 'battery life' or 'shipping delay', but it does not judge whether those topics were discussed favorably or unfavorably. Option C is incorrect because named entity recognition labels concrete entities like brand names, people, or locations mentioned in the text; it has no notion of tone and would not distinguish a glowing review from a scathing one. Option D is incorrect because language detection only identifies which natural language the review is written in, a preprocessing step that might run before analysis but tells the team nothing about customer satisfaction. Only sentiment analysis directly answers the question the team is asking: was this review positive or negative.