Skip to main content
Feature engineering turns raw data into inputs that improve model performance. Sourcetable automates common transformations through the AI assistant.

Missing value handling

Encoding categorical variables

Scaling and normalization

Feature creation

Date/time features

Creates: year, quarter, month, week, day of week, day of month, hour, minute, is_weekend, is_holiday, days_since_start.

Aggregation features

Group-by aggregations: sum, mean, median, count, min, max, std, first, last.

Interaction features

Products, ratios, and polynomial features between numeric columns.

Text features

Creates: word count, character count, average word length, sentiment score, TF-IDF vectors.

Feature selection

Importance-based

Trains a model and ranks features by their contribution to predictions.

Variance Inflation Factor (VIF)

Calculates VIF for each feature. Values above 5-10 indicate problematic multicollinearity.

Recursive Feature Elimination (RFE)

Iteratively removes the least important feature until performance stops improving.