Explain feature engineering and its importance in data science projects.

Feature engineering transforms raw data into meaningful variables that improve model performance and interpretability.\n\n• Techniques: Scaling, encoding categorical variables, creating interactions, polynomial features\n• Domain knowledge: Business-specific transformations and derived metrics\n• Automated methods: Feature selection algorithms, dimensionality reduction\n• Validation: Cross-validation to prevent overfitting and data leakage\n\nExample: E-commerce dataset enhancement includes creating recency/frequency/monetary features from transaction history, encoding product categories, generating time-based features (day of week, seasonality), and interaction terms between user demographics and purchase behavior.