AIF-C01 Question 204
Select 3A company is building a machine learning model for predicting customer churn and wants to ensure transparency and accountability in their process. Which of the following AWS features can help the company document the data origins, transformations, and the model’s metadata effectively?
- A
AWS Glue Data Catalog
- B
Amazon SageMaker Model Cards
- C
Amazon Rekognition
- D
AWS CloudTrail
- E
Amazon SageMaker Data Wrangler
Show answer and explanation
Correct answers: A, B, E
Explanation
Transparency and accountability in machine learning require proper documentation of data origins (data lineage) and metadata about the models. AWS Glue Data Catalog helps track data assets and their lineage, while SageMaker Model Cards focus on documenting model details, including the data used. Data Wrangler aids in documenting data preparation steps. Together, these tools help ensure a robust documentation process for machine learning workflows.
- A. Correct.
AWS Glue Data Catalog helps maintain a centralized repository of metadata about data assets, including their origins (data lineage). This supports documenting data sources and transformations.
- B. Correct.
Amazon SageMaker Model Cards provide a structured way to document information about machine learning models, including data origins, performance metrics, and intended use cases.
- C. Incorrect.
Amazon Rekognition is an AI service for image and video analysis. While it provides insights into media content, it does not support documenting data origins or metadata.
- D. Incorrect.
AWS CloudTrail tracks API activity and resource changes in AWS accounts but is not designed for documenting data origins or metadata related to machine learning models.
- E. Correct.
Amazon SageMaker Data Wrangler simplifies data preparation and transformation steps, allowing users to document these steps for transparency and reproducibility.