An ML engineer must ensure a dataset complies with PII regulations and prevent SageMaker training instances from using any PII. What is the most operationally efficient solution?
Choose an answer
Tap an option to check your answer.
Correct answer: Call Amazon Comprehend DetectPiiEntities to redact PII, store the redacted data in an Amazon S3 bucket, and have SageMaker access that S3 bucket for training..
Why this is the answer
The most operationally efficient solution is to use Amazon Comprehend DetectPiiEntities to redact PII and store the redacted data in Amazon S3. SageMaker natively integrates with S3 for data storage, making this a straightforward and scalable approach for training. Storing data in Amazon EFS (Elastic File System) and mounting it to SageMaker instances is less operationally efficient than S3. While EFS can be used, S3 is the standard and most integrated data source for SageMaker training jobs. AWS Glue DataBrew can cleanse data, but Comprehend is specifically designed for PII detection and redaction, making it a more direct tool for the task. Amazon Macie is excellent for discovering PII but does not directly offer PII removal capabilities; it focuses on identification and reporting.
Pass your exam — without the endless answer hunt
Get every verified question and explanation for this exam in one place, and save hours of prep. 1,000+ certifications · 20+ languages · free to start.
Pass your exam faster → No card needed