MLA-C01 Question 68
Single answerYou are working as a machine learning engineer for a retail company. You need to preprocess and clean a dataset containing customer purchase histories, which includes handling missing values, standardizing numerical features, and visualizing data distributions. Additionally, you want to create feature transformations and export the final dataset for training a predictive model in Amazon SageMaker. Which AWS service would best suit these requirements?
- A
Amazon SageMaker Data Wrangler
- B
AWS Glue
- C
AWS Glue DataBrew
- D
Amazon QuickSight
Show answer and explanation
Correct answer: A
Explanation
Amazon SageMaker Data Wrangler is the most appropriate tool for this scenario as it is purpose-built for machine learning workflows. It provides an intuitive user interface to clean, visualize, and transform data while offering integration with SageMaker for seamless model training. While AWS Glue and AWS Glue DataBrew are capable of data transformation, they are not as tailored for ML-specific preprocessing and feature engineering workflows. Amazon QuickSight, on the other hand, is unrelated to data preparation tasks.
- A. Correct.
Amazon SageMaker Data Wrangler is specifically designed for machine learning workflows. It provides tools to preprocess, clean, visualize, and transform data, as well as integrate seamlessly with SageMaker for model training.
- B. Incorrect.
AWS Glue is primarily a serverless ETL (Extract, Transform, Load) tool designed for large-scale data integration tasks. While it can process and transform data, it is not optimized for machine learning-specific workflows like feature engineering.
- C. Incorrect.
AWS Glue DataBrew is a visual data preparation tool that allows users to clean and transform data. However, it is not as tightly integrated with SageMaker as Data Wrangler and lacks advanced ML-specific preprocessing capabilities.
- D. Incorrect.
Amazon QuickSight is a business intelligence tool for generating reports and dashboards. It is not intended for data preprocessing or feature engineering tasks.