AI engineering glossary · Feature engineering data prep
What is feature crossing?
Feature crossing (also called Feature cross): Creating a new categorical feature from the combination of two or more categorical values, so each combination gets its own weight. It is common in linear models for ads and recommendations, where specific pairings carry signal.
Feature crossing: an example
Crossing city with day_of_week gives values like Mumbai_Friday and Pune_Sunday, each with its own learned effect.
Related terms
More Feature engineering data prep terms
- Feature engineering
- Feature extraction
- Feature scaling
- Feature selection
- Feature store
- Feature transformation
- Filter method
- Fit-transform pattern
Where you learn it
Feature crossing is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.