DAS-C01 · Question #140
A hospital is building a research data lake to ingest data from electronic health records (EHR) systems from multiple hospitals and clinics. The EHR systems are independent of each other and do not…
The correct answer is A. An AWS Glue ETL job with the FindMatches transform. The challenge is entity resolution - matching patient records across systems that have no common identifier - without requiring ML expertise. AWS Glue's FindMatches ML transform is purpose-built for this: it uses machine learning to identify records that refer to the same…
Question
A hospital is building a research data lake to ingest data from electronic health records (EHR) systems from multiple hospitals and clinics. The EHR systems are independent of each other and do not have a common patient identifier. The data engineering team is not experienced in machine learning (ML) and has been asked to generate a unique patient identifier for the ingested records. Which solution will accomplish this task?
Options
- AAn AWS Glue ETL job with the FindMatches transform
- BAmazon Kendra
- CAmazon SageMaker Ground Truth
- DAn AWS Glue ETL job with the ResolveChoice transform
How the community answered
(53 responses)- A79% (42)
- B6% (3)
- C4% (2)
- D11% (6)
Explanation
The challenge is entity resolution - matching patient records across systems that have no common identifier - without requiring ML expertise. AWS Glue's FindMatches ML transform is purpose-built for this: it uses machine learning to identify records that refer to the same entity across datasets, even with inconsistent naming, typos, or missing fields. Critically, FindMatches handles the ML complexity internally; the data engineering team only needs to provide labeled examples or let it learn from the data, requiring no ML experience. Option B (Amazon Kendra) is an enterprise search service. Option C (SageMaker Ground Truth) is a data labeling service. Option D (ResolveChoice) resolves ambiguous data type conflicts in Glue schemas, not record matching.
Topics
Community Discussion
No community discussion yet for this question.