Objective 1.1 Identify the Project Scope CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 2)

Questions 6–10

1. 6application · medium

   A marketing agency is developing a customer segmentation model for a client. The client has specified that the model must be delivered as a Power BI dashboard and that the segmentation should be based on purchasing behavior. The agency's data science team has proposed using a clustering algorithm. Which project specification is most critical to define before starting the work?

   • AThe specific clustering algorithm to use, such as K-means or DBSCAN.
   • BThe programming language to be used for the clustering implementation.
   • CThe type of cloud service to host the data pipeline.
   • DThe number of segments the client expects to see in the dashboard.
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A telecom company is developing a model to predict customer churn. The project sponsor wants the model to be deployed within 2 months, but the data science team has only 1 month of historical customer data, which may not capture seasonal patterns. The team is asked to define the project specifications. What should the team do?

   • APromise to deliver the model in 2 months with high accuracy, using the available data.
   • BAsk the sponsor to extend the deadline to 6 months to collect more data, without discussing alternatives.
   • CDefine the success metric as model precision and assume the data is sufficient.
   • DSet a realistic success metric, such as improving retention by 5%, and flag the data limitation to the sponsor, suggesting a phased approach.
   Reveal Solution Discussion Select an answer first
4. 8expert · hard

   A logistics company is building a model to optimize delivery routes. The project has a 2-month timeline. The team has access to GPS data from delivery vehicles, but the data is noisy and incomplete. The company also has a constraint that the model must run on the existing on-premises servers, which have limited memory. The project sponsor wants the model to reduce fuel costs by 10%. Which limitation is most likely to impact the model's ability to meet the sponsor's expectation?

   • AThe limited memory on the on-premises servers, which restricts the model's complexity.
   • BThe sponsor's expectation of a 10% fuel cost reduction, which is unrealistic for any model.
   • CThe 2-month timeline, which is too short to build any useful model.
   • DThe noisy and incomplete GPS data, which may limit the model's accuracy and ability to achieve the 10% fuel cost reduction.
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A government agency is building a model to predict tax evasion. The project has a 6-month timeline. The agency has access to a large dataset of tax returns, but the data is sensitive and cannot leave the agency's secure environment. The data science team has a dedicated on-premises cluster for training. The sponsor wants a model that can be used by auditors in the field, who have limited connectivity. Which deliverables should be classified as mandatory?

   • AA model that runs on the on-premises cluster, and a detailed technical report.
   • BA cloud-based API for real-time predictions, and a dashboard for auditors.
   • CA mobile app for auditors, and a real-time API for the agency's headquarters.
   • DA model that can run on auditors' laptops with limited connectivity, and a report on model performance.
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A manufacturing company wants to build a predictive maintenance model for its equipment. The data science team has access to sensor data from only 10% of the machines, and the rest are older models without sensors. The project sponsor expects the model to cover all machines. What is the most important limitation to communicate to the sponsor?

   • AThe model will be less accurate for older machines, but it can still be applied to all machines.
   • BThe model can only be developed for the machines with sensor data, and the scope must be adjusted accordingly.
   • CThe model can be built using only the available data, and the sponsor should not worry about the rest.
   • DThe project timeline must be extended to install sensors on all machines before modeling.
   Reveal Solution Discussion Select an answer first