A dataset contains a 'purchase_date' column. The data scientist wants to create features like 'day_of_week', 'month', and 'year' from this column. Which feature engineering technique should they use?

Convert dates to useful features

What is a potential drawback of hash encoding?

It creates too many columns for high-cardinality features.

It preserves the ordinal relationship between categories.

It can cause collisions, where different categories map to the same column.

It requires the target variable to be numeric.

What is the primary purpose of using smoothing in target encoding?

To reduce the number of columns created

To convert categories into binary columns

To ensure all categories have the same encoded value

To prevent overfitting by shrinking estimates for categories with few samples

What is the primary difference between dummy encoding and effect encoding?

Effect encoding uses the target variable, while dummy encoding does not.

Effect encoding represents the reference category with -1, while dummy encoding uses 0.

Effect encoding creates k columns, while dummy encoding creates k-1.

Effect encoding is only used for ordinal data, while dummy encoding is for nominal data.

CertNexusCertified Data Science Practitioner (CDSP)

Domain 3Objective 3

Objective 3.3 Carry Out Feature Engineering CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 9)

Part of the 3.0 Performing exploratory data analysis domain, which accounts for 25-36% of the CERTIFIED-DATA-SCIENCE-PRACTITIONER exam.

54questions here

11free pages

27concepts

25-36%of the exam

27 Concepts in This Objective

One-hot encoding
Target encoding
Label/Ordinal encoding
Dummy encoding
Effect encoding
Binary encoding
Base-N encoding
Hash encoding
Split features
Text manipulation - Split
Text manipulation - Trim
Text manipulation - Reverse
Manipulate data - Split names
Manipulate data - Extract year from title
Convert dates to useful features
PCA (Principal Component Analysis)
Missing value ratio
t-SNE (t-Distributed Stochastic Neighbor Embedding)
Low-variance filter
Random forest feature importance
High-correlation filter
Backward feature elimination
SVD (Singular Value Decomposition)
Forward feature selection
False discovery rate
Factor analysis
Feature importance methods

Questions 41–45

41foundation · easy
A dataset contains a 'purchase_date' column. The data scientist wants to create features like 'day_of_week', 'month', and 'year' from this column. Which feature engineering technique should they use?
Select an answer first
42foundation · easy
What is a potential drawback of hash encoding?
Select an answer first
43foundation · easy
What is the primary purpose of using smoothing in target encoding?
Select an answer first
44application · easy
A data scientist is working with a dataset that has several features with a high percentage of missing values. The dataset has 50,000 rows and 100 features. The goal is to reduce noise and improve model performance. Which feature selection method is most appropriate?
Select an answer first
45foundation · easy
What is the primary difference between dummy encoding and effect encoding?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to CERTIFIED-DATA-SCIENCE-PRACTITIONER

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 CertNexus. “CERTIFIED-DATA-SCIENCE-PRACTITIONER” is a trademark of its owner, used for identification only.