Problem
Scenario
You are working on a supervised learning problem and need to improve model quality beyond basic raw inputs. You want a structured way to create, validate, and ship features without overfitting or introducing leakage.
Question
How do you approach feature engineering in a machine learning project?
Representative Dataset
size·1.2M security-account-day rows, 62 raw featurestarget·Margin call within 5 trading daysfeatures·Portfolio, exposure, cash, volatility, segment, region, timestampsmissing_data·12% overall, concentrated in metadata and sparse event fieldsclass_balance·6.5% positive
You are practicing as a guest. Sign up free to get your answer graded with AI feedback. Your draft stays right here.
