In the model training process, what is the primary purpose of splitting data into training, validation, and test sets?

To evaluate the model's performance on data it has not seen during training or hyperparameter tuning

To ensure the model has enough data to learn all patterns perfectly

To reduce the size of the dataset for faster training

To allow the model to train on multiple subsets and average the results

A team is building a churn prediction model with 200 features. The model shows high variance on the validation set. They want to reduce overfitting while maintaining interpretability. Which approach is the best first step?

Apply feature selection to reduce the number of features.

Remove the validation set and train on all data.

Use a more complex algorithm, such as a deep neural network.

Add more features to capture more patterns.

A data scientist is choosing between two models. Model A has low bias but high variance. Model B has high bias but low variance. The goal is to minimize the total expected error on new data. What should they do?

Combine both models by averaging their predictions.

Always choose Model A because low bias is more important.

Always choose Model B because low variance is more important.

Choose the model with the best performance on a held-out test set.

SnowflakeSnowPro Advanced — Data Scientist

Domain 1Objective 1

Define Machine Learning Concepts for Data Science Workloads SNOWPRO-ADVANCED-DATA-SCIENTIST Practice Questions (Page 3)

Part of the Data Science Concepts domain, which makes up ~16% of our current practice bank.

24questions here

5free pages

6concepts

Questions 11–15

11foundation · easy
In the model training process, what is the primary purpose of splitting data into training, validation, and test sets?
Select an answer first
12expert · hard
A team is building a churn prediction model with 200 features. The model shows high variance on the validation set. They want to reduce overfitting while maintaining interpretability. Which approach is the best first step?
Select an answer first
13foundation · easy
Which machine learning approach is best suited for an agent that learns to play a game by receiving rewards for winning moves and penalties for losing moves?
Select an answer first
14expert · hard
A data scientist is choosing between two models. Model A has low bias but high variance. Model B has high bias but low variance. The goal is to minimize the total expected error on new data. What should they do?
Select an answer first
15foundation · easy
A data scientist wants to segment customers into distinct groups based on purchasing behavior without pre-defined labels. Which algorithm is most appropriate?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to SNOWPRO-ADVANCED-DATA-SCIENTIST

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 Snowflake. “SNOWPRO-ADVANCED-DATA-SCIENTIST” is a trademark of its owner, used for identification only.

Define Machine Learning Concepts for Data Science Workloads SNOWPRO-ADVANCED-DATA-SCIENTIST Practice Questions (Page 3)

Questions 11–15

In the model training process, what is the primary purpose of splitting data into training, validation, and test sets?

A team is building a churn prediction model with 200 features. The model shows high variance on the validation set. They want to reduce overfitting while maintaining interpretability. Which approach is the best first step?

Which machine learning approach is best suited for an agent that learns to play a game by receiving rewards for winning moves and penalties for losing moves?

A data scientist is choosing between two models. Model A has low bias but high variance. Model B has high bias but low variance. The goal is to minimize the total expected error on new data. What should they do?

A data scientist wants to segment customers into distinct groups based on purchasing behavior without pre-defined labels. Which algorithm is most appropriate?

Keep studying — Data Science Concepts