DA Data Cleaning and Preparation 2 — Questions and Answers
Question 1: What is data parsing in the context of data preparation?
- Sorting rows in ascending order
- Extracting structured information from raw or unstructured text (Correct answer)
- Merging two datasets
- Computing summary statistics
Correct answer: Extracting structured information from raw or unstructured text
Parsing breaks down raw data (like strings or JSON) into structured, usable components for analysis.
Question 2: Which technique rescales data so the mean is 0 and standard deviation is 1?
- Min-max normalization
- Log transformation
- Standardization (Z-score scaling) (Correct answer)
- Binning
Correct answer: Standardization (Z-score scaling)
Standardization subtracts the mean and divides by the standard deviation, centering the data at zero with unit variance.
Question 3: What is the purpose of data profiling?
- Generating charts for stakeholders
- Assessing data quality, structure, and content before analysis (Correct answer)
- Training predictive models
- Encrypting sensitive fields
Correct answer: Assessing data quality, structure, and content before analysis
Data profiling examines datasets to understand their structure, completeness, distributions, and anomalies before analysis begins.
Question 4: What is binning (bucketing) in data preparation?
- Removing rows with missing values
- Grouping continuous values into discrete categories or intervals (Correct answer)
- Encoding binary outcomes
- Splitting strings by delimiter
Correct answer: Grouping continuous values into discrete categories or intervals
Binning converts continuous numeric values into categorical bins, which can reduce noise and simplify analysis.
Question 5: What is a regex (regular expression) commonly used for in data cleaning?
- Joining tables on matching keys
- Pattern-based text extraction, validation, and replacement (Correct answer)
- Computing running totals
- Converting wide data to long format
Correct answer: Pattern-based text extraction, validation, and replacement
Regular expressions define patterns that match character sequences, enabling extraction, validation, or replacement of text in data fields.
Question 6: What does 'tidy data' mean as defined by Hadley Wickham?
- Data sorted by date
- Each variable is a column, each observation is a row, and each type of observational unit is a table (Correct answer)
- Data with no missing values
- Data stored in a single flat file
Correct answer: Each variable is a column, each observation is a row, and each type of observational unit is a table
Tidy data is structured so that each column is a variable, each row is an observation, and each table holds one observational unit, making it easy to work with.
DA Data Cleaning and Preparation 2
What is data parsing in the context of data preparation?