AWS Certified Solutions Architect – Associate (SAA-C03) — Question 760

A company needs to extract the names of ingredients from recipe records that are stored as text files in an Amazon S3 bucket. A web application will use the ingredient names to query an Amazon DynamoDB table and determine a nutrition score.

The application can handle non-food records and errors. The company does not have any employees who have machine learning knowledge to develop this solution.

Which solution will meet these requirements MOST cost-effectively?

Answer options

• A. Use S3 Event Notifications to invoke an AWS Lambda function when PutObject requests occur. Program the Lambda function to analyze the object and extract the ingredient names by using Amazon Comprehend. Store the Amazon Comprehend output in the DynamoDB table.
• B. Use an Amazon EventBridge rule to invoke an AWS Lambda function when PutObject requests occur. Program the Lambda function to analyze the object by using Amazon Forecast to extract the ingredient names. Store the Forecast output in the DynamoDB table.
• C. Use S3 Event Notifications to invoke an AWS Lambda function when PutObject requests occur. Use Amazon Polly to create audio recordings of the recipe records. Save the audio files in the S3 bucket. Use Amazon Simple Notification Service (Amazon SNS) to send a URL as a message to employees. Instruct the employees to listen to the audio files and calculate the nutrition score. Store the ingredient names in the DynamoDB table.
• D. Use an Amazon EventBridge rule to invoke an AWS Lambda function when a PutObject request occurs. Program the Lambda function to analyze the object and extract the ingredient names by using Amazon SageMaker. Store the inference output from the SageMaker endpoint in the DynamoDB table.

Correct answer: A

Explanation

Amazon Comprehend is a pre-trained natural language processing (NLP) service that can extract key phrases and entities from text files without requiring any machine learning expertise. Amazon Forecast is used for time-series forecasting and Amazon Polly is for text-to-speech, rendering options B and C incorrect. Amazon SageMaker requires machine learning expertise to build, train, and deploy models, and running a continuous SageMaker endpoint is not cost-effective for this use case.