A company has a large, unstructured dataset. The dataset includes many duplicate records across several key attributes. Which solution on AWS will detect duplicates in the dataset with the LEAST code

Question

Options

• AUse Amazon Mechanical Turk jobs to detect duplicates.
• BUse Amazon QuickSight ML Insights to build a custom deduplication model.
• CUse Amazon SageMaker Data Wrangler to pre-process and detect duplicates.
• DUse the AWS Glue FindMatches transform to detect duplicates.

• A
  3% (1)
• B
  6% (2)
• C
  13% (4)
• D
  78% (25)

Explanation

AWS Glue FindMatches is a built-in machine learning transform designed specifically for record linkage and deduplication - it requires no custom model code, just configuration through the Glue console or a simple API call, making it the lowest-effort solution for detecting duplicates across key attributes in unstructured data.

• A (Mechanical Turk) relies on human workers to manually review records - this is slow, expensive, and requires significant orchestration code to wire up, not a least-code solution.
• B (QuickSight ML Insights) is a BI/visualization tool; it does not offer a deduplication or entity-matching capability - this is a distractor leveraging the word "ML."
• C (SageMaker Data Wrangler) can preprocess data and has some duplicate-detection features, but building a usable deduplication workflow still requires substantially more configuration and code than Glue FindMatches.

Memory tip: Think "FindMatches = Find Matches = Find Duplicates" - the name literally describes the job, and it lives in Glue (the AWS data integration service), so when you see "deduplicate with least effort," glue those two words together: Glue FindMatches.

No community discussion yet for this question.