CLF-C02 Practice Question 66: Text Sentiment Analysis
AWS Certified Cloud Practitioner (CLF-C02) · Test 7 · question 6 of 10
An online store collects thousands of written product reviews every day. It wants to label each review automatically as positive, negative, neutral or mixed, so the product teams can spot problems quickly. No one at the company has machine learning experience. Which AWS service should it use?
Show answer and explanation
Answer:
Amazon Comprehend uses natural language processing (NLP) to extract insights from text: entities, key phrases, language, PII and sentiment. Its pre-trained models return the dominant sentiment of a document as positive, negative, neutral or mixed. AWS gives exactly this example: send each customer comment to the DetectSentiment operation to find out how customers feel about a product. No training data or textual analysis expertise is needed.
Option B, Amazon Rekognition, analyzes *images and videos*, for example to detect objects, faces or inappropriate content. It does not read the text of a review. Option C, Amazon Transcribe, converts *speech* into text. It would be a first step only if the reviews were audio recordings. Option D, Amazon Textract, extracts text and data from *scanned documents* and forms. It gets the words off a page but does not judge their sentiment.
Reference: What is Amazon Comprehend?