Data Management Controls AAISM Practice Questions (Page 6)

Questions 26–30

1. 26application · medium

   An AI system for fraud detection ingests transaction data, processes it in real time, and stores derived features for model retraining. The compliance team requires that raw transaction data be kept for 7 years, but derived features are only needed for 2 years. Which data lifecycle control should be implemented?

   • AAnonymize raw data after 2 years and keep derived features indefinitely.
   • BDelete raw data after 2 years and keep derived features for 7 years.
   • CApply the same retention period to both raw data and derived features to simplify compliance.
   • DRetain raw data for 7 years and derived features for 2 years, then securely delete the derived features.
   Reveal Solution Discussion Select an answer first
2. 27application · medium

   A company is developing an AI system for hiring. The data includes resumes, interview scores, and demographic information. The company wants to ensure that the data is used ethically and in compliance with regulations. Which governance structure should be established?

   • AA data ethics committee that reviews AI use cases and data handling practices.
   • BA data lineage team that documents data flow.
   • CA data security team that manages access controls.
   • DA data quality team that monitors data accuracy.
   Reveal Solution Discussion Select an answer first
4. 28application · medium

   A retail company is building a recommendation engine using customer purchase history. The data science team notices that the model performs poorly for new customers because their purchase history is sparse. The data quality team is asked to improve model performance. Which data quality dimension should they focus on?

   • ACompleteness
   • BTimeliness
   • CAccuracy
   • DConsistency
   Reveal Solution Discussion Select an answer first
5. 29application · medium

   A logistics company uses an AI model to predict delivery delays. The model is trained on data from multiple regional offices, but the data formats differ (e.g., dates as MM/DD/YYYY vs. DD/MM/YYYY). The model's accuracy is lower than expected. Which data quality issue is most likely the cause?

   • AInconsistency in data formats across sources.
   • BData lineage is not documented.
   • CInaccurate data entry in some offices.
   • DIncomplete data for some regions.
   Reveal Solution Discussion Select an answer first
6. 30application · medium

   A government agency uses AI to process citizen applications. The data includes sensitive personal information. The agency must ensure that only authorized personnel can access the data and that all access is logged. Which control is most directly aligned with this requirement?

   • AStore the data in a secure data center.
   • BImplement role-based access control (RBAC) and enable audit logging.
   • CEncrypt the data at rest and in transit.
   • DAnonymize the data before processing.
   Reveal Solution Discussion Select an answer first