Data Science Project Lifecycle Stages Quiz

Submit

3. True or False: Feature scaling is unnecessary when using tree-based models.

True

False

4. Which stakeholders should be involved in the problem definition stage of a data science project?

Data scientists only

Business teams, domain experts, and data scientists

IT operations only

Executive management only

Submit

6. What is the purpose of A/B testing in model deployment?

To compare two model versions in production

To split training data

To reduce model complexity

To handle missing values

7. Which of the following are important metrics for assessing model fairness? (Select all that apply)

Demographic parity

Equalized odds

Precision only

Disparate impact ratio

Submit

9. True or False: A model with high training accuracy and low test accuracy indicates overfitting.

True

False

10. Which validation technique is most suitable for time-series data?

K-fold cross-validation

Time-series cross-validation

Stratified sampling

Random holdout

11. Which stage of the data science lifecycle involves identifying business problems and defining measurable objectives?

Problem Definition

Data Collection

Model Evaluation

Deployment

12. Version control and documentation are critical during which lifecycle stage?

Problem Definition only

Development and Deployment

All stages of the lifecycle

Evaluation only

13. What does 'data drift' refer to in the context of deployed models?

Gradual changes in input data distribution over time

Errors in data collection

Missing values in datasets

Imbalanced class labels

14. True or False: Once a model is deployed to production, no further modifications are needed.

True

False

15. Which of the following are key responsibilities during the model monitoring phase? (Select all that apply)

Track model prediction drift

Rewrite original code

Monitor input data quality

Update model when performance degrades

Submit

17. In model evaluation, which metric is most appropriate for imbalanced classification problems?

Accuracy

F1-Score

Mean Squared Error

R-squared

18. What is the primary purpose of train-test splitting in model development?

To reduce computational cost

To evaluate model performance on unseen data

To increase dataset size

To balance class distribution

19. True or False: Exploratory Data Analysis (EDA) should occur before data cleaning.

True

False

Submit

DataX Data Science Project Lifecycle Stages Quiz

1. The process of selecting relevant features for model training is called ____.

2.

What first name or nickname would you like us to use?

2. In the operations phase, SLA (Service Level Agreement) monitoring ensures ____.

3. True or False: Feature scaling is unnecessary when using tree-based models.

4. Which stakeholders should be involved in the problem definition stage of a data science project?

5. Model ____ refers to the practice of regularly retraining models with new data.

6. What is the purpose of A/B testing in model deployment?

7. Which of the following are important metrics for assessing model fairness? (Select all that apply)

8. In the deployment phase, containerization using Docker helps achieve ____.

9. True or False: A model with high training accuracy and low test accuracy indicates overfitting.

10. Which validation technique is most suitable for time-series data?

11. Which stage of the data science lifecycle involves identifying business problems and defining measurable objectives?

12. Version control and documentation are critical during which lifecycle stage?

13. What does 'data drift' refer to in the context of deployed models?

14. True or False: Once a model is deployed to production, no further modifications are needed.

15. Which of the following are key responsibilities during the model monitoring phase? (Select all that apply)

16. Hyperparameter tuning involves adjusting model parameters to ____.

17. In model evaluation, which metric is most appropriate for imbalanced classification problems?

18. What is the primary purpose of train-test splitting in model development?

19. True or False: Exploratory Data Analysis (EDA) should occur before data cleaning.

20. Data preprocessing typically includes cleaning, handling missing values, and ____.