Feature engineering means creating and choosing the inputs a model uses. Better features often improve a model more than a more complex algorithm does.
Common examples
- From a date, extract the month, the weekday or whether it is a holiday.
- From a purchase history, calculate the number of orders in the last 90 days.
- Turn categories, such as city, into numbers a model can use.
- Combine two columns, such as price per square metre.
Scale and clean
Some models work better when numbers are on a similar scale. Handle missing values in a planned way, and record what you did.
Avoid leakage
Never use information that would not be available at the time of the prediction, such as a column filled in after the outcome is known. It makes a model look excellent in testing and fail in real use.
Next step: see our data science course at ScholarLogic, or book a free demo class. A counsellor will share the fee and the next batch dates.
