4.2 Describe Automated Data Preparation Tasks (quality Checks, Formatting, Transformation, Cleaning) AI-TECHNICAL-PRACTITIONER Practice Questions (Page 2)

Questions 6–10

1. 6expert · hard

   A data team is automating the data preparation for a dataset that will be used to train a regression model. The dataset contains a 'height' field with some values that are clearly impossible (e.g., 300 cm) and a 'weight' field with some missing values. The team wants to ensure data quality without introducing bias. Which automated approach should they implement?

   • AReplace impossible height values with the mean height and fill missing weight with zero.
   • BRemove all records with impossible height values and drop records with missing weight.
   • COnly flag impossible height values; missing weight values are not a data quality issue.
   • DFlag impossible height values for review, and impute missing weight values using the median weight for similar height groups.
   Reveal Solution Discussion Select an answer first
2. 7expert · hard

   A data engineer is automating the cleaning of a dataset containing customer feedback. The dataset has a 'rating' field (1-5) with some values of 0, a 'comment' field with many missing entries, and a 'date' field with inconsistent formats. The team wants to prepare the data for sentiment analysis, which requires text data and may not need the rating field. Which automated cleaning strategy should they implement?

   • AOnly standardize the date format; rating and comments are not important for sentiment analysis.
   • BReplace rating 0 with the average rating and leave comments and dates as is.
   • CRemove all records with missing comments and keep rating 0 values.
   • DRemove records with rating 0, impute missing comments with an empty string, and standardize the date format.
   Reveal Solution Discussion Select an answer first
4. 8expert · hard

   A data team is building an automated pipeline for a sensor network. The sensors occasionally produce readings that are physically impossible (e.g., temperature of -300°C) and sometimes fail to report (missing values). The team wants to ensure data quality without losing too much data. Which automated approach should they implement?

   • AReplace all readings outside the plausible range with the mean and fill missing values with zero.
   • BOnly flag readings outside the plausible range; missing values are not a quality issue.
   • CRemove all readings outside the plausible range and drop records with missing values.
   • DFlag readings outside a plausible range (e.g., -50°C to 50°C) for review, and impute missing values using the last known reading.
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A data analyst is preparing a dataset for a machine learning model that predicts customer churn. The dataset contains a 'signup_date' field in various formats and a 'total_spend' field with some negative values (which are invalid). They also need to create a new feature 'months_since_signup'. Which automated data preparation steps should they implement?

   • ARemove records with negative 'total_spend' and calculate 'months_since_signup' without standardizing dates.
   • BStandardize 'signup_date' and calculate 'months_since_signup', but keep negative 'total_spend' values.
   • CStandardize 'signup_date' to a common format, remove records with negative 'total_spend', and calculate 'months_since_signup' from the standardized date.
   • DOnly standardize 'signup_date'; negative spend and feature creation are not data preparation tasks.
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   A data pipeline automatically flags values that fall far outside the expected range for a numeric column. Which type of quality check is being performed?

   • AFormat standardization
   • BDuplicate detection
   • COutlier detection
   • DData aggregation
   Reveal Solution Discussion Select an answer first