What does a learning curve plot typically show?

Model performance (e.g., training and validation scores) as a function of training set size

The distribution of classes in the dataset

The number of parameters in the model over time

The loss function value over epochs during training

A data science team built a binary classifier to flag fraudulent transactions. The business cost of missing a fraud (false negative) is very high, while false positives are merely an inconvenience. The team evaluates the model on a test set and obtains a confusion matrix with 950 true negatives, 30 false positives, 15 false negatives, and 5 true positives. Which metric should the team prioritize when deciding whether to deploy this model?

Accuracy, because it reflects the overall proportion of correct predictions.

Recall, because it measures the proportion of actual frauds correctly identified.

Precision, because it measures how many flagged transactions are truly fraudulent.

F1-score, because it balances precision and recall into a single number.

What is the primary purpose of Kubeflow?

To orchestrate machine learning workflows on Kubernetes

To replace Python for model training

To provide a spreadsheet interface for data analysis

To visualize data with dashboards

Which of the following is a key feature of Kubeflow for model deployment?

It only supports deep learning models

It eliminates the need for containerization

It provides scalable serving of models using Kubernetes

It automatically tunes hyperparameters for all models

What is the purpose of serializing a trained model (e.g., using pickle or joblib) in Python?

To automatically retrain the model on new data

To convert the model into a format that can be saved to disk and loaded later for inference

To visualize the model's decision boundaries

To improve the model's accuracy by compressing it

CertNexusCertified Data Science Practitioner (CDSP)

Domain 4Objective 3

Objective 4.3 Evaluate Models 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.

30questions here

6free pages

8concepts

19-27%of the exam

Questions 21–25

21foundation · easy
What does a learning curve plot typically show?
Select an answer first
22application · medium
A data science team built a binary classifier to flag fraudulent transactions. The business cost of missing a fraud (false negative) is very high, while false positives are merely an inconvenience. The team evaluates the model on a test set and obtains a confusion matrix with 950 true negatives, 30 false positives, 15 false negatives, and 5 true positives. Which metric should the team prioritize when deciding whether to deploy this model?
Select an answer first
23foundation · easy
What is the primary purpose of Kubeflow?
Select an answer first
24foundation · easy
Which of the following is a key feature of Kubeflow for model deployment?
Select an answer first
25foundation · easy
What is the purpose of serializing a trained model (e.g., using pickle or joblib) in Python?
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.