Remove Outliers from a Spark DataFrame Based on Standard Deviation or IQR MACHINE-LEARNING-ASSOCIATE Practice Questions (Page 1)

Questions 1–4

1. 1foundation · easy

   A data scientist computes Q1 = 10 and Q3 = 30 for a numeric column. The IQR is 20. Which lower and upper bounds should be used to remove outliers using the standard 1.5 * IQR rule?

   • ALower = 10, Upper = 30
   • BLower = -10, Upper = 50
   • CLower = 0, Upper = 40
   • DLower = -20, Upper = 60
   Reveal Solution Discussion Select an answer first
2. 2foundation · easy

   Which Spark DataFrame method is used to exclude rows that do not meet a specified condition?

   • Aagg()
   • Bselect()
   • Cfilter()
   • Ddrop()
   Reveal Solution Discussion Select an answer first
4. 3foundation · easy

   Which Spark SQL function is used inside a window function to compute the first quartile (Q1) of a numeric column?

   • Aavg()
   • Bpercentile_approx()
   • Ccount()
   • Dstddev()
   Reveal Solution Discussion Select an answer first
5. 4foundation · easy

   A data scientist wants to remove rows from a Spark DataFrame where the value in column 'amount' is more than 3 standard deviations away from the mean. Which approach correctly computes the necessary statistics and filters the DataFrame?

   • ACompute the mean and standard deviation with agg(), then filter using a column expression comparing 'amount' to the bounds.
   • BUse the drop() method with the column name to remove outlier rows.
   • CUse the approxQuantile() method to get the mean and standard deviation, then filter with a SQL expression.
   • DApply the stddev() function directly to the filtered DataFrame to remove outliers.
   Reveal Solution Discussion Select an answer first