Tim recognized noise from misrepresentation of data in large data collections and decided to apply a data-quality technique to prepare the data for analysis. Which technique did he rely on?

• A. Data normalization
• B. Data reduction
• C. Data visualization
• D. Data archiving

Answer: (B)

When data is large and messy, the goal is to keep the important information while stripping away the extra, often noisy, or redundant parts. Data reduction achieves this by shrinking the data representation to the most informative features or summaries. By removing redundant attributes and focusing on the key signals, the dataset becomes easier to analyze and less susceptible to distortion from misrepresented measurements. This improves the reliability of downstream analyses and speeds up processing, which is especially valuable with big data.

Normalization is about making features comparable in scale, which helps certain algorithms but doesn’t by itself remove misrepresentation or noise. Data visualization helps explore the data but isn’t primarily a preparation technique to clean or compress data. Archiving stores data for long-term retention and doesn’t address data quality either.