Which method is commonly used to handle missing values by substituting a statistic from the data?

• A. Normalization
• B. Discretization
• C. Binning
• D. Imputation

Answer: (D)

Imputation is the technique used to handle missing values by replacing them with a statistic derived from the observed data. By filling in the gaps with a value such as the mean, median, or mode of the feature, you preserve the dataset size and enable standard algorithms to run without special handling for missing entries. This approach relies on information already present in the data to provide a plausible estimate for the missing value. Other methods like normalization or discretization/binning serve different purposes (scaling data or converting continuous values into categories) and don’t fill in missing data themselves. Advanced imputation can use models or multiple imputation to better reflect uncertainty in the estimates.