# What is Feature Engineering? Definition & Technical Mechanics

> **Source:** [SPIDITS AI AI Glossary](https://spidits.com/ai-glossary)  
> **Category:** Model Training  
> **Canonical Citation:** [https://spidits.com/ai-glossary/feature-engineering](https://spidits.com/ai-glossary/feature-engineering)

## Definition
Feature Engineering is the process of using domain knowledge to select, transform, combine, and manipulate raw variables into highly predictive input features for machine learning algorithms.

## Architectural Mechanics & Deep Dive
Feature engineering is the process of using domain knowledge to select, transform, combine, and clean raw data variables to create highly informative features for machine learning models. Effective feature engineering improves model accuracy and reduces computational complexity, involving operations like creating interaction terms, bucketizing continuous variables, and extracting date parts.

## Industry Applications & Real-World Use Cases
Tabular machine learning datasets, model regression preprocessing, and feature selection runs.

## Frequently Asked Questions

### Why is feature engineering important in classical machine learning?
Because algorithms like linear regression, decision trees, and SVMs cannot automatically discover complex relationships between variables. Creating meaningful features manually directly improves model accuracy and training efficiency.

### Give examples of feature engineering transformations.
One-hot encoding of categorical variables, normalising numerical coordinates, extracting hour-of-day from timestamp strings, and creating ratios of two related variables.

## Related Terminology
- [feature](https://spidits.com/ai-glossary/feature)
- [one-hot-encoding](https://spidits.com/ai-glossary/one-hot-encoding)
- [data-preprocessing](https://spidits.com/ai-glossary/data-preprocessing)

---
*From the SPIDITS AI Knowledge Base (336+ terms indexed). Explore full technical definitions: [https://spidits.com/ai-glossary](https://spidits.com/ai-glossary)*