What is feature normalization using Z-score standardization?

Updated May 16, 2026

Short answer

Z-score standardization scales features to have mean 0 and standard deviation 1.

Deep explanation

Z-score transformation makes features comparable by centering them around zero and scaling based on variance. This is essential for gradient-based models and distance-based algorithms.

Real-world example

Used in credit scoring models to standardize income, age, and debt ratios.

Common mistakes

  • Applying scaling before train-test split leading to leakage.

Follow-up questions

  • What is Z-score formula?
  • When should Z-score not be used?

More Feature Engineering interview questions

View all →