Removing columns from a data set because they are not useful for a certain type of data analysis is an example of _________.

• A. Data cleaning
• B. Data transformation
• C. Normalization
• D. Attribute reduction

Answer: (D)

Removing columns that aren’t useful for a specific analysis targets reducing the number of attributes we consider, which is known as attribute reduction. This is a form of feature selection: we keep only the informative attributes and drop the rest. By pruning irrelevant or redundant columns, the model becomes simpler, runs faster, and often generalizes better because it’s less affected by noise.

This approach is different from data cleaning (fixing errors or missing values), data transformation (changing the representation of data, like encoding or scaling), and normalization (scaling features to a common range). The core idea here is to streamline the feature set by removing unnecessary columns to focus on the information that matters for the analysis.