An ML engineer needs to ensure that a dataset complies with regulations for personally identifiable information (PII). The ML engineer will use the data to train an ML model on Amazon SageMaker…

Question

Options

• AUse the Amazon Comprehend DetectPiiEntities API call to redact the PII from the data. Store the
• BUse the Amazon Comprehend DetectPiiEntities API call to redact the PII from the data. Store the
• CUse AWS Glue DataBrew to cleanse the dataset of PII. Store the data in an Amazon Elastic File
• DUse Amazon Macie for automatic discovery of PII in the data. Remove the PII. Store the data in

• A
  60% (12)
• B
  25% (5)
• C
  5% (1)
• D
  10% (2)

Explanation

Option A is correct because Amazon Comprehend's DetectPiiEntities API is purpose-built for identifying and redacting PII from text at scale, and storing the cleansed data in Amazon S3 pairs naturally with SageMaker - which treats S3 as its native data source, making the pipeline seamless and requiring minimal custom infrastructure.

Why the distractors fail:

• Option B likely stores the redacted data in a less SageMaker-native location (e.g., EFS or RDS), adding unnecessary complexity compared to S3.
• Option C (Glue DataBrew) is a general-purpose data preparation tool; it can handle PII but requires more configuration and isn't optimized for PII redaction the way Comprehend is - less efficient for this specific task.
• Option D (Amazon Macie) is a discovery and alerting service, not a redaction tool. It flags where PII exists in S3 but doesn't remove it automatically, requiring additional manual or scripted steps to actually remediate.

Memory tip: Think of it this way - Macie = finds PII, Comprehend = fixes PII. For exam questions asking about removing or redacting PII efficiently in an ML pipeline, Comprehend's DetectPiiEntities (or ContainsPiiEntities) API is the go-to answer, especially when SageMaker + S3 are involved.

No community discussion yet for this question.