Artificial Intelligence & Machine Learning
Feature Engineering
Definition
Feature engineering is the process of using domain knowledge to extract features (characteristics, properties, attributes) from raw data. It involves creating new features or selecting the most relevant ones to improve the performance of a machine learning model.
Why It Matters
Feature engineering is often the most time-consuming but also the most impactful part of a machine learning project. Better features often lead to better models, even with simpler algorithms.
Contextual Example
From a simple `date` column, you could engineer new features like "day of the week," "month," and "is_holiday." These new features might be much more predictive for a sales forecasting model than the original date alone.
Common Misunderstandings
- It is a combination of art and science, requiring both creativity and domain expertise.
- Deep learning models can automate some aspects of feature engineering by learning features directly from raw data.