Describe the Data Scientist Role and Related Skills DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 5)

Questions 21–24

1. 21expert · hard

   A ride-sharing company collects real-time location data from millions of drivers and riders. The data science team wants to build a model that predicts demand in different areas to optimize driver allocation. The data is streaming and extremely large. Which approach best aligns with the data scientist's role in this big data context?

   • ABuild a simple rule-based system that sends more drivers to areas with high historical demand.
   • BHave a team of analysts manually monitor live maps and dispatch drivers based on their judgment.
   • CUse a distributed streaming platform (e.g., Kafka) and a scalable machine learning framework (e.g., Spark MLlib) to train and deploy a real-time demand prediction model.
   • DStore all the data in a relational database and run a batch prediction model every night.
   Reveal Solution Discussion Select an answer first
2. 22foundation · easy

   What is the primary responsibility of a data scientist within an organization?

   • ATo design and maintain the data storage systems and databases
   • BTo manage the IT infrastructure and ensure network security
   • CTo create visual dashboards and reports for business stakeholders
   • DTo extract insights from data and communicate them to drive business decisions
   Reveal Solution Discussion Select an answer first
4. 23expert · hard

   A data scientist at a bank has built a credit risk model that is highly accurate. However, regulators require that the bank be able to explain why a loan was denied to each applicant. The data scientist is struggling to meet this requirement because the model is a complex ensemble. Which approach best addresses the regulatory need while maintaining model performance?

   • AUse model-agnostic explanation techniques like SHAP or LIME to provide per-prediction explanations.
   • BAsk the business analyst to write a manual explanation for each denied loan.
   • CReplace the ensemble with a simple logistic regression model, even if accuracy drops slightly.
   • DHire a data engineer to build a new data pipeline that captures more features for the model.
   Reveal Solution Discussion Select an answer first
5. 24expert · hard

   A streaming service collects viewing data from millions of users. The data science team wants to build a recommendation engine that updates in near real-time as users watch content. The team currently uses a batch processing system that updates recommendations nightly. Which change is most aligned with the data scientist's role in a big data context?

   • AUse a larger cluster to run the nightly batch faster.
   • BImplement a stream processing framework (e.g., Apache Flink) to update recommendations incrementally as new viewing events occur.
   • CKeep the nightly batch system but increase the frequency to every hour.
   • DHave a team of analysts manually curate recommendations for popular content.
   Reveal Solution Discussion Select an answer first