Oversee Data Evaluation PMI-CPMAI Practice Questions (Page 4)

Questions 16–20

1. 16application · medium

   A data analyst is performing exploratory data analysis (EDA) on a customer segmentation dataset. The analyst creates a histogram of customer age and notices a sharp spike at age 0. The dataset documentation indicates that age 0 is used as a placeholder for missing values. What is the most appropriate next step in the EDA process?

   • AReplace age 0 with the mean age of the dataset.
   • BInvestigate the records with age 0 to determine if they should be treated as missing and decide on an imputation strategy.
   • CIgnore the spike because it is a small fraction of the data.
   • DRemove all records with age 0 immediately to avoid skewing the distribution.
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   Which of the following is a common technique used in exploratory data analysis?

   • ASetting up a load balancer
   • BWriting SQL queries to delete data
   • CCreating histograms to visualize distributions
   • DConfiguring a firewall
   Reveal Solution Discussion Select an answer first
4. 18application · medium

   During exploratory data analysis of a marketing response dataset, a data scientist creates a correlation matrix and finds that two features, 'total_emails_sent' and 'total_emails_opened', have a correlation coefficient of 0.98. What should the data scientist do next in the EDA process?

   • ARemove both features immediately to avoid multicollinearity.
   • BIgnore the correlation because it is a natural relationship.
   • CConvert both features to categorical variables.
   • DInvestigate the relationship further and consider whether one feature is redundant for modeling.
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   A retail company is developing a demand forecasting model for seasonal products. The available historical sales data spans the last 10 years, but the company's product catalog has changed significantly over that period, with many products discontinued and new ones introduced. The team must decide whether to use the full 10-year dataset or only the last 2 years. Which data evaluation criterion is most critical in this decision?

   • AConsistency
   • BRelevance
   • CCompleteness
   • DAccuracy
   Reveal Solution Discussion Select an answer first
6. 20application · medium

   A data scientist is analyzing a dataset of employee salaries for a pay equity model. The histogram of salaries shows a right-skewed distribution, with a long tail of high salaries. The model will be used to predict salaries based on job role and experience. What should the data scientist consider during EDA?

   • AThe data is incomplete because it lacks salaries below a certain threshold.
   • BThe skewness may affect model performance, so consider applying a log transformation to the target variable.
   • CThe high salaries are outliers and should be removed.
   • DThe distribution is normal, so no action is needed.
   Reveal Solution Discussion Select an answer first