Objective 4.2 Train Models CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 4)

Questions 16–20

1. 16application · medium

   A data science team has a small dataset and wants to evaluate the performance of a classification model. They are concerned about the variance of the performance estimate. Which evaluation method is most appropriate?

   • ATraining on the entire dataset and evaluating on the same data
   • BA single train-test split
   • CUsing a holdout validation set only
   • DK-fold cross-validation
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   What is the main advantage of Bayesian optimization over grid search for hyperparameter tuning?

   • AIt requires no hyperparameters to be specified
   • BIt can only be used for deep learning models
   • CIt is more efficient by using past evaluations to inform future searches
   • DIt is guaranteed to find the global optimum
   Reveal Solution Discussion Select an answer first
4. 18foundation · easy

   What is the key assumption made by a naïve Bayes classifier?

   • AThe data is linearly separable
   • BFeatures are independent given the class label
   • CThe target variable is continuous
   • DFeatures are normally distributed
   Reveal Solution Discussion Select an answer first
5. 19foundation · easy

   What is the typical use case for a transformer model like BERT?

   • ADimensionality reduction
   • BClustering unlabeled data
   • CNatural language processing tasks
   • DTime series forecasting
   Reveal Solution Discussion Select an answer first
6. 20application · easy

   A spam filter needs to classify emails as spam or not spam based on the frequency of certain words. The team has a large dataset and needs a fast, simple model that works well with high-dimensional text data. Which model is most appropriate?

   • ALinear regression
   • BK-means clustering
   • CNaïve Bayes
   • DARIMA
   Reveal Solution Discussion Select an answer first