Data Wrangling DY0-001 Practice Questions (Page 3)

Questions 11–15

1. 11expert · hard

   A data science team is building a model to classify support tickets into categories. They have a large unlabeled dataset and a small set of expert-labeled tickets. The team wants to minimize labeling effort while maximizing model performance. They are considering active learning. Which approach is most effective?

   • AUse the model's uncertainty to select tickets where the model is least confident for expert labeling.
   • BUse clustering to select a diverse set of tickets for expert labeling.
   • CRandomly select tickets for expert labeling until the budget is exhausted.
   • DLabel all tickets using the model's predictions without expert review.
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   Which imputation strategy uses the most frequently occurring value in a column to fill missing data points?

   • AMean imputation
   • BMedian imputation
   • CRegression imputation
   • DMode imputation
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A data engineer needs to combine two datasets: 'orders' (order_id, customer_id, order_date, amount) and 'customers' (customer_id, name, city). The goal is to create a dataset that includes all orders, even those from customers not present in the 'customers' table, while still including customer information when available. Which merge strategy should the engineer use?

   • ARight join on customer_id
   • BFull outer join on customer_id
   • CLeft join on customer_id
   • DInner join on customer_id
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A data analyst is merging two datasets: one contains user IDs as integers, the other contains user IDs as strings with leading zeros (e.g., '00123'). After merging, the analyst notices that many users appear duplicated. What is the most likely cause and the best solution?

   • AThe merge key has different data types; convert both columns to the same type (e.g., string) before merging.
   • BThe merge is using an inner join; switch to a left join to avoid duplicates.
   • CThe datasets contain duplicate rows; remove duplicates from each dataset before merging.
   • DThe datasets have different column names; rename the columns to match before merging.
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A healthcare analytics team is building a model to predict patient readmission. The dataset has a 'blood_pressure' column with 15% missing values, and the missingness is related to the patient's age (older patients are more likely to have missing readings). The team wants to minimize bias in the imputed values. Which imputation approach is most appropriate?

   • AListwise deletion of rows with missing blood pressure
   • BMultiple imputation by chained equations (MICE)
   • CReplace missing values with zero
   • DMean imputation
   Reveal Solution Discussion Select an answer first