AWS Machine Learning Engineer Associate (MLA-C01) · 28% practice allocation

AWS MLA-C01: Data Preparation for Machine Learning Practice Questions

5 free sample questions from the Data Preparation for Machine Learning domain, with answers and explanations. The full MLA-C01 bank has 59 questions across 4 domains. Questions are scenario-based and written from the official AWS exam guide.

1.

A visual, no-code tool is needed for data analysts to clean and normalize data with 250+ prebuilt transformations, without writing Spark code. Which service fits?

  • A. Amazon Inspector
  • B. AWS Glue DataBrew✓ Correct
  • C. AWS Direct Connect
  • D. Amazon SageMaker Model Registry
Explanation

AWS Glue DataBrew is a visual, no-code data-preparation tool with many prebuilt transformations for cleaning and normalizing data. Model Registry (D) versions models, Inspector (A) scans for security vulnerabilities, and Direct Connect (C) is network connectivity — none is a no-code data-prep tool.

2.

Which file format is generally BEST for storing large ML training datasets in Amazon S3 for efficient, columnar analytical reads?

  • A. Apache Parquet✓ Correct
  • B. Plain uncompressed CSV
  • C. A single large JSON document
  • D. A Word document
Explanation

Parquet is columnar and compressed, giving efficient reads and smaller storage for large datasets — well-suited to ML pipelines. Uncompressed CSV (B) is row-based and larger, a single JSON doc (C) is unwieldy at scale, and a Word document (D) is not a data format for ML.

3.

A dataset has a categorical column 'color' with values red/green/blue that must be fed to an algorithm requiring numeric input, with no implied ordering. Which encoding is appropriate?

  • A. Dropping the column
  • B. Normalizing the strings
  • C. Simply assigning red=1, green=2, blue=3 (label/ordinal encoding)
  • D. One-hot encoding✓ Correct
Explanation

One-hot encoding creates a binary column per category, avoiding a false ordinal relationship. Assigning 1/2/3 (C) implies an ordering the colors don't have, which can mislead many algorithms. Dropping the column (A) discards signal, and 'normalizing strings' (B) isn't a valid numeric encoding.

4.

A pipeline ingests high-velocity streaming events (clickstream) that must be captured continuously for near-real-time feature computation. Which AWS service is designed for streaming ingestion?

  • A. Amazon Kinesis Data Streams✓ Correct
  • B. AWS Artifact
  • C. Amazon Comprehend
  • D. Amazon S3 Glacier Deep Archive
Explanation

Amazon Kinesis Data Streams ingests high-throughput streaming data in real time for downstream processing. Glacier Deep Archive (D) is cold storage, Artifact (B) is compliance docs, and Comprehend (C) is an NLP service — none handles streaming ingestion.

5.

A serverless ETL job must catalog, clean, and transform data from multiple sources before ML training, using Spark under the hood. Which service fits BEST?

  • A. AWS Shield
  • B. Amazon EC2 with manual Spark installation only
  • C. AWS Glue✓ Correct
  • D. Amazon CloudFront
Explanation

AWS Glue is a serverless ETL service (with a data catalog and managed Spark) for preparing and transforming data. Manually installing Spark on EC2 (B) is not serverless and adds ops burden; CloudFront (D) is a CDN and Shield (A) is DDoS protection — neither does ETL.

Test yourself across every MLA-C01 domain

The free MLA-C01 practice test mixes about 20 questions weighted like the exam. The full 59-question bank adds instant grading, flashcards and a timed mock exam.

Take the free MLA-C01 practice testPractice the full bank

Study this domain

Data Preparation for Machine Learning study guide (MLA-C01) →

Other MLA-C01 domains

← All AWS Machine Learning Engineer Associate study resources