Which action is a valid treatment for outliers after they have been detected?

Capping the outlier values at a certain percentile

Ignoring the outliers without any analysis

Duplicating the outlier rows to increase their weight

Converting all outliers to NULL

What is the purpose of partitioning data into training and testing sets?

To make the data easier to visualize

To evaluate model performance on unseen data

To ensure all data is used for training

To reduce the size of the dataset for storage

A data analyst discovers that a 'product_code' column contains values like 'A-123', 'a-123', and 'A123' which should all refer to the same product. The analyst also finds that some rows have a 'product_code' of 'UNKNOWN' or are NULL. The business wants to standardize the product codes for reporting, but also wants to track which rows had data quality issues. What is the best approach?

Keep the data as-is and rely on the reporting tool to handle the inconsistencies.

Normalize the product_code by uppercasing and removing hyphens, and create a flag column for rows that were originally 'UNKNOWN' or NULL.

Remove all rows with 'UNKNOWN' or NULL product codes.

Replace all 'UNKNOWN' and NULL values with a single value 'MISC' and normalize the rest.

What is the main difference between exact and fuzzy matching for deduplication?

Exact matching is used for small datasets, while fuzzy matching is used for large datasets

Exact matching is used for numeric data, while fuzzy matching is used for text data

Exact matching requires identical values, while fuzzy matching allows for minor variations

Exact matching is faster but less accurate than fuzzy matching

Snowflake SnowPro Advanced:Data Analyst

Domain 4Objective 2

Prepare and Clean Datasets SNOWPRO-ADVANCED-DATA-ANALYST Practice Questions (Page 7)

Part of the Data Ingestion and Data Preparation domain, which accounts for 17% of the SNOWPRO-ADVANCED-DATA-ANALYST exam.

43questions here

9free pages

15concepts

17%of the exam

Questions 31–35

31foundation · easy
Which action is a valid treatment for outliers after they have been detected?
Select an answer first
32foundation · easy
What is the purpose of partitioning data into training and testing sets?
Select an answer first
33expert · hard
A data analyst discovers that a 'product_code' column contains values like 'A-123', 'a-123', and 'A123' which should all refer to the same product. The analyst also finds that some rows have a 'product_code' of 'UNKNOWN' or are NULL. The business wants to standardize the product codes for reporting, but also wants to track which rows had data quality issues. What is the best approach?
Select an answer first
34foundation · easy
What is the main difference between exact and fuzzy matching for deduplication?
Select an answer first
35foundation · easy
Which function would you use to extract the year from a date column?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to SNOWPRO-ADVANCED-DATA-ANALYST

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Snowflake. “SNOWPRO-ADVANCED-DATA-ANALYST” is a trademark of its owner, used for identification only.

Prepare and Clean Datasets SNOWPRO-ADVANCED-DATA-ANALYST Practice Questions (Page 7)

Questions 31–35

Which action is a valid treatment for outliers after they have been detected?

What is the purpose of partitioning data into training and testing sets?

What is the main difference between exact and fuzzy matching for deduplication?

Which function would you use to extract the year from a date column?

Keep studying — Data Ingestion and Data Preparation