Describe Best Practices for Communicating Findings and Operationalizing an Analytics Project DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 3)

Questions 11–15

1. 11expert · hard

   A company has a model in production that predicts customer churn. The model's performance has been stable, but the business team wants to change the definition of 'churn' from 'no purchase in 90 days' to 'no purchase in 60 days'. What is the most important consideration before making this change?

   • AThe model's monitoring dashboard should be updated to reflect the new definition, but the model itself does not need to change.
   • BThe change should be rejected because it would make the model's historical performance metrics incomparable.
   • CThe model will need to be retrained with the new target definition, and its performance must be re-evaluated against the new business objective.
   • DThe model's predictions can be used as-is, since the change is only in the business definition and does not affect the model's input data.
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A model that predicts equipment failure has been in production for six months. The data science team notices that the model's precision has dropped significantly over the last month, even though the input data distribution appears unchanged. What is the most appropriate first step to diagnose this issue?

   • AImmediately retrain the model on the most recent data to improve its performance.
   • BIncrease the frequency of model retraining to daily to keep the model current.
   • CRoll back to the previous version of the model that had better precision.
   • DReview the model's prediction logs and compare them against actual outcomes to identify where the errors are occurring.
   Reveal Solution Discussion Select an answer first
4. 13foundation · easy

   When presenting analytical findings to a group of executives, what is the most important factor to consider?

   • AThe statistical significance of every variable
   • BThe amount of data used in the analysis
   • CThe technical complexity of the model
   • DThe business impact and strategic implications
   Reveal Solution Discussion Select an answer first
5. 14foundation · easy

   What is a key practice for governance of an operationalized analytics solution?

   • AKeeping the model's code secret from stakeholders
   • BAvoiding any updates to the model
   • CDeleting logs after each prediction
   • DDocumenting model versions and changes
   Reveal Solution Discussion Select an answer first
6. 15foundation · easy

   What is the first step in deploying an analytical model into a production environment?

   • ADeleting the development environment
   • BDefining the deployment requirements and success criteria
   • CTraining the model on all available data
   • DWriting the final report
   Reveal Solution Discussion Select an answer first