AI engineering glossary · Feature engineering data prep
What is max-abs scaling?
Max-abs scaling (also called MaxAbsScaler): A scaling method that divides each feature by its largest absolute value so results fall between -1 and 1. It does not shift the data, so zeros stay zeros, which makes it suitable for sparse data where preserving sparsity matters.
Max-abs scaling: an example
A sparse column with values 0, 0, 4, -8 becomes 0, 0, 0.5, -1 after dividing by 8, and the zeros stay exactly zero.
Related terms
More Feature engineering data prep terms
- Mean imputation
- Median imputation
- Min-max scaling
- Missing at random
- Missing completely at random
- Missing indicator
- Missing not at random
- Missing value imputation
Where you learn it
Max-abs scaling is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.