AWS Certified Machine Learning Engineer Associate · Free Practice Question Medium
Question 30
A travel agency has collected a large volume of customer feedback recordings from calls made to its customer support team after launching a new vacation package. The agency wants to evaluate the success of the vacation package by analyzing the sentiment of customer feedback. The ML engineer needs to process and analyze the recordings to identify positive and negative sentiments in the least amount of time.
Which action should the ML engineer take?
-
A
Use Amazon Transcribe to convert the customer feedback recordings into text, and then use Amazon Comprehend to analyze the text for sentiment
-
B
Use AWS Glue to preprocess the feedback recordings and implement a sentiment analysis solution using open-source libraries
-
C
Manually transcribe the recordings, convert them to English using Amazon Translate, and run a custom script to analyze sentiment
-
D
Build a custom sentiment analysis model using Amazon SageMaker and train it with labeled customer feedback data
Reveal correct answer
Correct answer: A
Explanation
Correct option:
Use Amazon Transcribe to convert the customer feedback recordings into text, and then use Amazon Comprehend to analyze the text for sentiment
This solution uses AWS services optimized for the workflow:
Amazon Transcribe - Converts audio feedback recordings into text, enabling text-based analysis.
Amazon Comprehend - Analyzes text for sentiment (positive, negative, neutral, mixed), providing actionable insights.
This solution leverages fully managed AWS services that minimize the need for custom coding and manual preprocessing. Also, the integration of Transcribe and Comprehend enables efficient and scalable sentiment analysis with minimal operational overhead.
Incorrect options:
Build a custom sentiment analysis model using Amazon SageMaker and train it with labeled customer feedback data - Training a custom model requires significant time and effort to prepare the data, train the model, and validate its accuracy. Managed services like Comprehend eliminate this overhead for sentiment analysis.
Manually transcribe the recordings, convert them to English using Amazon Translate, and run a custom script to analyze sentiment - Manually transcribing the recordings introduces unnecessary delays. Additionally, Amazon Comprehend supports multiple languages natively, making translation unnecessary.
Use AWS Glue to preprocess the feedback recordings and implement a sentiment analysis solution using open-source libraries - Amazon Polly converts text to speech, not speech to text. Writing a custom script for sentiment analysis adds complexity compared to using Amazon Comprehend.
References:
https://aws.amazon.com/transcribe/
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
