Ingest and Transform Data DP-203 Practice Questions (Page 7)

Questions 31–35

1. 31foundation · easy

   What is a common strategy for handling missing data in a dataset?

   • AIgnoring missing values and proceeding with analysis
   • BImputing missing values with the mean or median of the column
   • CReplacing missing values with random numbers
   • DDeleting the entire dataset if any value is missing
   Reveal Solution Discussion Select an answer first
2. 32foundation · easy

   Why is denormalization often used in analytical workloads?

   • ATo improve query performance by reducing the number of joins
   • BTo reduce data redundancy and improve write performance
   • CTo enforce referential integrity
   • DTo make data more secure
   Reveal Solution Discussion Select an answer first
4. 33application · medium

   A data science team uses Azure Synapse Analytics with Apache Spark to analyze user interaction logs. The logs contain duplicate entries due to retries in the data collection system. You need to identify and remove duplicates before performing exploratory data analysis. The duplicate detection should be based on a combination of user ID and timestamp. What should you do?

   • AUse the Spark DataFrame API to group by user ID and timestamp, and count occurrences to identify duplicates.
   • BUse the dropDuplicates() method on the DataFrame specifying the columns 'user_id' and 'timestamp'.
   • CUse a SQL query with SELECT DISTINCT * FROM logs.
   • DUse the distinct() method on the entire DataFrame.
   Reveal Solution Discussion Select an answer first
5. 34application · medium

   A telecommunications company receives call detail records (CDRs) in a single large file. The records include a region code. You need to split the data into separate files by region for downstream processing. You are using Apache Spark in Azure Synapse Analytics. What should you do?

   • AUse the repartition function on the region column and write the DataFrame to a single file.
   • BUse the partitionBy function on the region column when writing the DataFrame.
   • CUse the groupBy function on the region column and write the aggregated result.
   • DUse the filter function to create a separate DataFrame for each region and write each one individually.
   Reveal Solution Discussion Select an answer first
6. 35application · medium

   A data science team has a large dataset of customer transactions in Azure Data Lake Storage Gen2. Before building a model, they need to understand the distribution of transaction amounts, identify outliers, and check for missing values. You need to perform exploratory data analysis using Apache Spark. What should you do?

   • AUse the describe function on the DataFrame to get summary statistics, then use the count and isNull functions to check for missing values.
   • BUse the sample function to take a random sample and then plot it with matplotlib.
   • CUse the groupBy function to aggregate transaction amounts by customer ID.
   • DUse the dropDuplicates function to remove duplicate rows and then count the remaining rows.
   Reveal Solution Discussion Select an answer first