What is the purpose of setting a fixed random seed when splitting a dataset into training and test sets?

To guarantee the model achieves the same accuracy on every run

To ensure the same split is generated each time the code is run

A credit card fraud detection dataset has 95% legitimate transactions and 5% fraudulent ones. The team plans to split the data into training (70%) and test (30%) sets. Which splitting method should they use to ensure the test set reflects the original class distribution?

Temporal split by transaction date

Split by customer ID to avoid leakage

Random split without stratification

Stratified split based on the target class

A team is building a model to predict daily electricity demand using historical data from the past three years. They need to evaluate the model's performance on the most recent year. Which splitting approach is most appropriate?

Temporal split: train on the first two years, validate on the next six months, test on the last six months

Split by month, randomly assigning months to training or test

Random split of all days into training and test sets

Stratified split by day of the week

For a time-series dataset, why is it important to split the data in chronological order rather than randomly?

To avoid using future data to predict past events, which would give unrealistic performance estimates

To ensure the training and test sets have the same class distribution

To increase the size of the training set

To make the model more robust to outliers

When is it appropriate to use non-random (e.g., temporal or chronological) splitting instead of random splitting?

When the data is time-series and the goal is to predict future values

When the goal is to maximize the training set size

CertNexusCertified Data Science Practitioner (CDSP)

Domain 4Objective 1

Objective 4.1 Prepare Data Sets for Modeling CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 2)

Part of the 4.0 Building models domain, which accounts for 19-27% of the CERTIFIED-DATA-SCIENCE-PRACTITIONER exam.

26questions here

6free pages

7concepts

19-27%of the exam

Questions 6–10

6foundation · easy
What is the purpose of setting a fixed random seed when splitting a dataset into training and test sets?
Select an answer first
7application · medium
A credit card fraud detection dataset has 95% legitimate transactions and 5% fraudulent ones. The team plans to split the data into training (70%) and test (30%) sets. Which splitting method should they use to ensure the test set reflects the original class distribution?
Select an answer first
8application · medium
A team is building a model to predict daily electricity demand using historical data from the past three years. They need to evaluate the model's performance on the most recent year. Which splitting approach is most appropriate?
Select an answer first
9foundation · easy
For a time-series dataset, why is it important to split the data in chronological order rather than randomly?
Select an answer first
10foundation · easy
When is it appropriate to use non-random (e.g., temporal or chronological) splitting instead of random splitting?
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.