Model Building - Describe Details of This Phase, Including Activities and Associated Roles DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 1)

Questions 1–5

1. 1expert · hard

   A healthcare organization is building a model to predict patient readmission risk. The data science team includes a data scientist, a data engineer, and a domain expert (a clinician). The data engineer has prepared the data and set up the environment. The data scientist has trained a model, but the clinician notes that the model's predictions are not clinically meaningful. The team needs to refine the model. Which action should the team take?

   • AIncorporate the clinician's domain knowledge to guide feature selection and model interpretation
   • BAsk the data engineer to increase the compute resources to train a more complex model
   • CDeploy the model and let clinicians adjust their workflows to fit the model
   • DIgnore the clinician's feedback because the model is statistically valid
   Reveal Solution Discussion Select an answer first
2. 2foundation · easy

   Which activity is performed during the model building phase?

   • ACreating dashboards to visualize model results for stakeholders
   • BSelecting an appropriate algorithm for the data and problem
   • CDefining the business problem and success criteria
   • DCleaning and transforming raw data into a usable format
   Reveal Solution Discussion Select an answer first
4. 3foundation · easy

   Which role typically provides the prepared, clean datasets that the data scientist uses to train models?

   • ABusiness sponsor
   • BData engineer
   • CData scientist
   • DProject manager
   Reveal Solution Discussion Select an answer first
5. 4application · medium

   A marketing analytics team is building a churn prediction model. The data engineer has provisioned a cloud environment with the necessary libraries and frameworks. The data scientist is now iterating on feature engineering and model parameters. Which responsibility belongs to the data engineer during this phase?

   • ADefining the success metrics for the churn model
   • BInterpreting the model results for business stakeholders
   • CSelecting the final algorithm for the churn model
   • DEnsuring the compute environment and data pipelines are reliable and scalable
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A bank is building a credit risk model. The data scientist has selected a gradient boosting algorithm and is now training the model. The data engineer has set up the training cluster. What should the data scientist do next to ensure the model is reliable?

   • AValidate the model on a holdout dataset to check for overfitting
   • BUse the entire dataset for training to maximize accuracy
   • CAsk the data engineer to clean the training data again
   • DImmediately deploy the model to production to start scoring loans
   Reveal Solution Discussion Select an answer first