Oversee Data Evaluation PMI-CPMAI Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   A dataset has no missing values in any field. Which data quality dimension is being assessed as satisfactory?

   • AAccuracy
   • BCompleteness
   • CConsistency
   • DValidity
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A healthcare AI project is using patient records from multiple hospitals. Each hospital records 'blood_pressure' in different units: some use mmHg and others use kPa. The data team is merging these datasets for model training. Which data quality dimension is most directly compromised if this unit discrepancy is not resolved?

   • AConsistency
   • BFreshness
   • CCompleteness
   • DAccuracy
   Reveal Solution Discussion Select an answer first
4. 13foundation · easy

   What is the primary purpose of reviewing the data schema and structure during data evaluation?

   • ATo encrypt the data at rest
   • BTo increase the volume of data
   • CTo reduce the number of features
   • DTo ensure compatibility with modeling requirements
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A financial institution is developing a fraud detection model. The available transaction data is from the last 12 months, but fraud patterns change rapidly. The team is considering using only the most recent 3 months of data to train the model. What is the primary trade-off in this decision?

   • AImproved relevance to current fraud patterns versus reduced sample size and potential loss of rare fraud cases.
   • BImproved data accuracy versus increased risk of overfitting.
   • CReduced data schema complexity versus increased preprocessing time.
   • DIncreased data completeness versus decreased model interpretability.
   Reveal Solution Discussion Select an answer first
6. 15expert · hard

   A data scientist is performing EDA on a dataset for a predictive maintenance model. The dataset contains sensor readings from machines, and the target variable is 'failure' (binary). During EDA, the scientist finds that the 'failure' class is rare (2% of records). Additionally, the sensor readings show a bimodal distribution, suggesting two different operating modes. What is the most important consideration for the modeling phase?

   • AThe rare class imbalance and the bimodal distribution both need to be addressed to avoid a biased model.
   • BThe rare class is not a problem because the dataset is large.
   • CThe bimodal distribution indicates that the data is not accurate and should be discarded.
   • DThe sensor readings should be normalized to a standard normal distribution.
   Reveal Solution Discussion Select an answer first