Gather Required Data PMI-CPMAI Practice Questions (Page 4)

Questions 16–20

1. 16foundation · easy

   Which approach is best for implementing an ongoing data collection process that requires data to be pulled from a source every hour?

   • ASchedule a recurring automated job
   • BManually export data each hour
   • CUse a one-time bulk load
   • DSet up a real-time streaming pipeline
   Reveal Solution Discussion Select an answer first
2. 17application · medium

   A data science team is building a churn prediction model using customer data extracted from multiple source systems. During the extraction process, they notice that some customer records have missing values in the 'last_purchase_date' field, and others have duplicate entries. The team needs to ensure the dataset is complete and accurate before training. What should they do?

   • AImplement data validation checks in the extraction pipeline that flag missing values and duplicates, and then apply a deduplication step and impute missing dates using the customer's account creation date.
   • BRemove all records with missing values and duplicates from the dataset before training to ensure data quality.
   • CIgnore the missing values and duplicates because the model can handle them during training.
   • DManually review and correct each record with missing values or duplicates in the source systems.
   Reveal Solution Discussion Select an answer first
4. 18expert · hard

   A marketing analytics team is building a customer segmentation model using data from a CRM and a web analytics platform. The data is extracted daily, but the team has noticed that the number of records varies significantly from day to day, and some records have inconsistent formats for the 'country' field. The model's accuracy is suffering because of these data quality issues. What is the most effective way to address this?

   • AIgnore the data quality issues because the model can handle them during training.
   • BUse a machine learning model to automatically detect and correct data quality issues in the dataset.
   • CManually review the data each day after extraction and correct any issues before loading it into the model.
   • DImplement data quality rules in the extraction pipeline that validate record counts against expected ranges and standardize the 'country' field using a lookup table.
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   A healthcare organization is migrating patient records from an on-premises EHR system to an Azure Synapse Analytics workspace for AI model development. The data contains PHI and must be encrypted in transit and at rest. The migration must also verify that no records are lost or corrupted during the transfer. What should the team configure?

   • AUse Azure Data Factory with a self-hosted IR, enable TLS for the connection, and add a copy activity with a data consistency check that compares row counts between source and sink.
   • BUse AzCopy to copy the data directly from the on-premises server to Azure Blob Storage, then load it into Synapse using PolyBase.
   • CUse Azure Database Migration Service to migrate the data, which automatically encrypts all data and performs integrity checks.
   • DExport the data to a flat file, encrypt it with a customer-managed key, and upload it to Azure Storage using SFTP.
   Reveal Solution Discussion Select an answer first
6. 20foundation · easy

   A data engineer needs to extract data from a legacy mainframe system that does not provide a modern API. Which method is most appropriate for this extraction?

   • AUse a REST API call
   • BUse a flat file export (e.g., CSV) from the mainframe
   • CUse a web scraping tool
   • DUse a NoSQL query
   Reveal Solution Discussion Select an answer first