AI engineering glossary · Feature engineering data prep
What is mutual information?
Mutual information: A measure of how much knowing one variable reduces uncertainty about another, based on entropy. Unlike correlation it captures non-linear relationships, so it is used to rank features against the target.
Mutual information: an example
Mutual information between hour_of_day and traffic congestion is high even though their linear correlation is near zero because the pattern rises and falls.
Related terms
More Feature engineering data prep terms
- Nominal variable
- One-hot encoding
- Ordinal encoding
- Ordinal variable
- Outlier treatment
- Oversampling
- Polynomial features
- Power transform
Where you learn it
Mutual information is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.