Short Definition
Feature engineering is the process of transforming raw data into model inputs that make useful patterns easier to learn.
Intuition
It is the craft of deciding what information the model should see and how that information should be represented.
Technical Definition
Feature engineering includes selecting, transforming, encoding, normalizing, combining, or deriving variables before model training.
Example
Converting timestamps into day-of-week and hour-of-day features can help a demand forecasting model.
Common Misunderstandings
Deep learning reduced some manual feature engineering, but it did not eliminate representation choices.
Feature engineering can leak future information if time boundaries are ignored.