MLA-C01 Question 82
Select 3You are working on a machine learning project to predict customer churn using data stored in an Amazon S3 bucket. Upon inspecting the dataset, you notice missing values in multiple columns, inconsistent date formats, and duplicate records. Your goal is to ensure data integrity before proceeding with modeling. Which of the following steps should you take to address these issues?
- A
Use Amazon SageMaker Data Wrangler to identify and remove duplicate records from the dataset.
- B
Impute missing values using techniques such as mean, median, or mode depending on the column type.
- C
Use AWS Glue DataBrew to standardize the date formats across the dataset.
- D
Ignore the missing values and proceed with the dataset as it is.
- E
Write a custom Lambda function to drop duplicate records and normalize the data.
Show answer and explanation
Correct answers: A, B, C
Explanation
Ensuring data integrity and preparing data for modeling involves addressing issues such as missing values, inconsistent formats, and duplicate records. AWS services like SageMaker Data Wrangler and Glue DataBrew are specifically designed to handle these tasks efficiently. Ignoring data quality issues or relying on custom scripts when managed solutions are available is not recommended in most scenarios.
- A. Correct.
Correct: Amazon SageMaker Data Wrangler is a powerful tool for data preparation that allows you to identify and remove duplicate records easily without writing custom scripts.
- B. Correct.
Correct: Imputing missing values is a common practice in data preparation to ensure the dataset is complete and suitable for modeling.
- C. Correct.
Correct: AWS Glue DataBrew is designed for data cleaning tasks such as standardizing date formats, making it an effective choice for this issue.
- D. Incorrect.
Incorrect: Ignoring missing values can lead to poor model performance and unreliable predictions. Data integrity is essential for effective modeling.
- E. Incorrect.
Incorrect: While writing a custom Lambda function is possible, AWS services like SageMaker Data Wrangler and Glue DataBrew provide more efficient, purpose-built solutions for these tasks.