Feature engineering

How it is used

Often the part of a project that improves a model most. Feature engineering relies on domain knowledge; typical steps are ratios, time windows, encoding categories and aggregating events. Finished features are often kept in a 'feature store'.

Example sentences

1. Good feature engineering beat the fancy model: a simple ratio of returns to orders did most of the work.

2. Most of the feature engineering is done in SQL so that it runs in the warehouse.

Common collocations

• do feature engineering
• feature engineering step
• feature store
• domain knowledge

Related terms

• featureIn modelling, one of the input variables a model uses to make predictions, such as a customer's age, number of purchases or days since last login.
• modelA mathematical representation of a real process, built from data, that is used to explain relationships or to make predictions about new cases.
• variableA characteristic that can take different values from one case to another, such as age, city or price; in a table it is usually one column.
• aggregateTo combine many rows of data into summary values, such as totals, counts or averages, usually grouped by a category such as day, product or region.

This word in other fields

The same English word can have a different meaning in another field:

• feature engineering · Software Engineering

More terms: Modelling and Machine Learning

• training data
• overfitting
• regression
• classification
• accuracy
• precision
• recall
• cross-validation
• baseline
• target variable
• test set
• clustering