Besides using a fixed random seed, what other practice helps ensure reproducibility of data splitting?

Recording the exact splitting parameters (e.g., test size, stratification flag) in the project documentation

Performing the split only once and never repeating it

Using a different random seed for each experiment

Splitting the data manually in a spreadsheet

A data scientist is preparing a dataset of 10,000 records for a classification model. They want to allocate the majority of data to training, with a smaller portion for validation and the smallest portion for testing. Which split ratio best reflects this common practice?

70% training, 20% validation, 10% test

50% training, 30% validation, 20% test

60% training, 20% validation, 20% test

80% training, 10% validation, 10% test

Why is it important to keep the test set separate from the training and validation sets?

To speed up the training process

To reduce the risk of data leakage from the validation set

To allow the model to learn from more data

To provide an unbiased evaluation of the final model's performance

A medical dataset contains 1,000 samples with a rare disease (positive class) and 99,000 healthy samples. The team wants to split the data into training and test sets. Which approach is most appropriate to ensure the test set contains enough positive cases for meaningful evaluation?

Oversample the positive class before splitting

Stratified split ensuring the same 1% positive rate in both sets

Use a 50/50 split to ensure enough positive cases

Random split with a 90/10 ratio

CertNexusCertified Data Science Practitioner (CDSP)

Domain 4Objective 1

Objective 4.1 Prepare Data Sets for Modeling CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 5)

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 21–25

21foundation · easy
Besides using a fixed random seed, what other practice helps ensure reproducibility of data splitting?
Select an answer first
22foundation · easy
A data scientist is preparing a dataset of 10,000 records for a classification model. They want to allocate the majority of data to training, with a smaller portion for validation and the smallest portion for testing. Which split ratio best reflects this common practice?
Select an answer first
23foundation · easy
A dataset for a binary classification problem has 90% negative examples and 10% positive examples. Which splitting method is most appropriate to ensure the training and test sets both contain approximately 10% positive examples?
Select an answer first
24foundation · easy
Why is it important to keep the test set separate from the training and validation sets?
Select an answer first
25application · medium
A medical dataset contains 1,000 samples with a rare disease (positive class) and 99,000 healthy samples. The team wants to split the data into training and test sets. Which approach is most appropriate to ensure the test set contains enough positive cases for meaningful evaluation?
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.