Objective 4.2 Train Models CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 9)

Questions 41–43

1. 41expert · hard

   A data science team is building a random forest regression model to predict housing prices. They have a dataset with many features, some of which are irrelevant. They are concerned about overfitting and want to evaluate the model's generalization. Which approach is most effective?

   • AUse random forest with feature importance and k-fold cross-validation
   • BUse a single decision tree with a train-test split
   • CUse random forest without feature importance and a single train-test split
   • DUse linear regression with all features and a train-test split
   Reveal Solution Discussion Select an answer first
2. 42expert · hard

   A data science team is tuning a deep neural network with many hyperparameters. Each training run is computationally expensive, and they need to find good hyperparameters with as few runs as possible. Which approach is most appropriate?

   • ARandom search
   • BBayesian optimization
   • CManual tuning
   • DGrid search
   Reveal Solution Discussion Select an answer first
3. 43foundation · easy

   What do the 'AR' and 'MA' components in ARIMA stand for?

   • AAutoRegressive and Machine Average
   • BAutoRegression and Model Average
   • CAutoRegressive and Moving Average
   • DAverage Regression and Moving Average
   Reveal Solution Discussion Select an answer first