Describe Best Practices for Communicating Findings and Operationalizing an Analytics Project DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 4)

Questions 16–20

1. 16application · medium

   A data analyst has created a line chart showing sales trends over the past year. The chart is cluttered with too many data points and gridlines, making it hard for the sales team to see the overall trend. Which change best applies the principles of clear data visualization?

   • ASimplify the chart by removing unnecessary gridlines, reducing the number of data points shown, and adding a clear title and axis labels.
   • BReplace the line chart with a pie chart to show the sales trend over time.
   • CAdd more data points to the chart to provide a more complete picture of the sales data.
   • DUse a 3D effect on the chart to make it more visually appealing and engaging for the sales team.
   Reveal Solution Discussion Select an answer first
2. 17expert · hard

   A data scientist is creating a presentation for a mixed audience of technical and non-technical stakeholders. The goal is to communicate the results of a complex predictive model. Which approach best balances the needs of both groups?

   • AFocus the presentation on the business impact and avoid all technical details, since the non-technical audience is the majority.
   • BPresent all the technical details first, then summarize the main findings at the end.
   • CCreate two separate presentations: one for the technical audience and one for the non-technical audience.
   • DUse a high-level narrative with clear visuals for the main findings, and include a technical appendix with detailed methodology for those who want it.
   Reveal Solution Discussion Select an answer first
4. 18foundation · easy

   Why is ongoing monitoring of a deployed analytics model important?

   • ATo reduce the need for data collection
   • BTo ensure the model continues to perform accurately over time
   • CTo make the model's predictions more random
   • DTo increase the model's complexity
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   A bank is deploying a credit risk model. The model will be used to make lending decisions. Which practice is most important to include in the operationalization plan to ensure the model remains compliant and effective over time?

   • ADeploy the model and only update it when a major regulatory change occurs.
   • BFocus exclusively on maximizing the model's accuracy at deployment, since a highly accurate model will not need frequent updates.
   • CEnsure the model is deployed as a batch process that runs monthly, since this is the most common and reliable method for credit models.
   • DImplement a monitoring system that tracks model performance and data drift, with a defined process for periodic review and retraining.
   Reveal Solution Discussion Select an answer first
6. 20expert · hard

   A financial services firm is deploying a model to approve small business loans. The model must be explainable to regulators and customers. The data science team is considering two approaches: a highly accurate gradient boosting model with limited interpretability, and a slightly less accurate logistic regression model that is fully interpretable. Which approach best satisfies the firm's requirements?

   • ADeploy the gradient boosting model and use a separate explainability tool like SHAP to provide post-hoc explanations.
   • BDeploy the gradient boosting model because its higher accuracy will lead to better loan decisions and higher profitability.
   • CDeploy both models and use the gradient boosting model for internal decisions and the logistic regression model for customer-facing explanations.
   • DDeploy the logistic regression model because its interpretability is essential for regulatory compliance and customer explanations, even if it is slightly less accurate.
   Reveal Solution Discussion Select an answer first