Google Professional Data Engineer Question 23
Single answerGoogle Cloud PlatformYou are tasked with preparing a dataset for machine learning training. The dataset contains raw transactional data stored in a Cloud Storage bucket, and it includes missing values, duplicate records, and inconsistent date formats. You need to clean and prepare this data using a fully-managed service that supports data transformation and profiling without writing extensive code. Which tool should you use?
- A
Cloud Dataflow
- B
Cloud Dataprep
- C
Cloud Data Fusion
- D
BigQuery
Show answer and explanation
Correct answer: B
Explanation
Cloud Dataprep is the most suitable tool for this scenario because it is specifically designed for data preparation and cleaning. It provides a visual interface for profiling and transforming data, making it an ideal choice for users who want to clean and prepare datasets without writing extensive code. The other options either require more coding effort or are designed for different purposes (such as building pipelines or querying data).
- A. Incorrect.
Cloud Dataflow is a powerful tool for building scalable data processing pipelines, but it requires you to write code in Apache Beam or use templates. It is not the best choice for users looking for a low-code solution for data cleaning and preparation.
- B. Correct.
Cloud Dataprep is a fully-managed, visual data preparation tool that supports profiling, cleaning, and transformation of datasets. It is ideal for users who want to prepare and clean data without extensive coding.
- C. Incorrect.
Cloud Data Fusion is a fully-managed, code-optional data integration solution designed for building data pipelines. While it has transformation capabilities, it is more suited for integrating and orchestrating data flows than for detailed data cleaning and profiling.
- D. Incorrect.
BigQuery is a serverless, highly scalable data warehouse primarily used for querying and analyzing large datasets. It is not specifically designed for data cleaning or preparation tasks.