Discovery - Describe Details of This Phase, Including Activities and Associated Roles DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 4)

Questions 16–20

1. 16application · medium

   A government agency is initiating a data science project to improve public transportation scheduling. During Discovery, the team includes a data scientist, a data engineer, and a domain expert from the transportation department. The domain expert is concerned that the team does not understand the constraints of bus routes. What is the best way to address this concern during Discovery?

   • AHave the domain expert lead a workshop to explain the business context and constraints.
   • BRemove the domain expert from the team to avoid bias.
   • CAsk the data scientist to research bus route constraints online.
   • DProceed with the project and assume the data will reveal the constraints.
   Reveal Solution Discussion Select an answer first
2. 17application · medium

   A manufacturing company wants to predict equipment failures before they occur. During the Discovery phase, the data science team interviews plant managers and learns that sensor data is collected from machines, but some sensors are not calibrated consistently. The team also finds that maintenance logs are handwritten and not digitized. What should the team do during Discovery to address these findings?

   • AProceed with modeling using only the sensor data and ignore the maintenance logs.
   • BDigitize the maintenance logs immediately before completing the Discovery phase.
   • CDocument the data quality issues and identify alternative data sources or additional data collection needs.
   • DAsk the plant managers to manually enter maintenance data into a spreadsheet for the project.
   Reveal Solution Discussion Select an answer first
4. 18application · medium

   A retail company wants to reduce customer churn. During the Discovery phase, the data science team meets with business stakeholders and learns that the marketing department defines churn as a customer not purchasing for 90 days, while the finance department uses 180 days. The team also discovers that customer transaction data is stored in a data warehouse and social media engagement data is available from a third-party API. What is the most important activity the team should perform during Discovery to ensure the subsequent analytics phases are not wasted?

   • AImmediately begin cleaning the social media data because it is likely to be messy and unstructured.
   • BAsk the IT department to provide a copy of all customer data to the data science team for exploration.
   • CBuild a preliminary predictive model using the transaction data to see if churn can be predicted.
   • DFormalize a single, agreed-upon definition of churn and document it as part of the business problem framing.
   Reveal Solution Discussion Select an answer first
5. 19expert · hard

   A multinational corporation is starting a data science project to optimize its supply chain. The project involves data from multiple countries, each with different data privacy regulations. During Discovery, the team identifies that some data sources are subject to strict data residency requirements, meaning the data cannot leave the country of origin. The project sponsor wants to use a centralized data lake for analysis. What is the most important consideration for the Discovery phase?

   • AThe team should ignore the residency requirements and centralize all data in the data lake.
   • BThe team should ask the legal department to waive the residency requirements.
   • CThe team must document the data residency constraints and plan for a federated or decentralized data architecture.
   • DThe team should only use data from countries with no residency restrictions.
   Reveal Solution Discussion Select an answer first
6. 20foundation · easy

   Which of the following is a key activity in the Discovery phase?

   • ASelecting the final machine learning algorithm
   • BPerforming feature engineering
   • CIdentifying data sources
   • DMonitoring model performance in production
   Reveal Solution Discussion Select an answer first