Navigation: <-- Part III: Data Understanding | Main Index | Part V: 1st Pass - Supervised Learning -->

---

Part IV: Data Preparation

Raw data rarely arrives ready for modeling. This part covers foundational preparation steps: transforming and encoding attributes, handling missing values and structural errors. These operations run before training a model, but are often informed by both EDA insights AND your modeling intent. Therefore, this part sits between EDA and modeling.

---

Nuggets in This Part

#	Nugget	Prerequisites
1	Feature Engineering	Data Types and Measurement Scales · EDA: Correlations
2	Structural Cleaning and Encoding	Data Types and Measurement Scales · EDA: Data Quality
3	Scaling and Imputation	Structural Cleaning and Encoding · EDA: Descriptive Statistics
4	Data Splits	Supervised Learning
5	Data Preparation Checklist	--
6	Data Preparation Best Practices	--

---

Script v1.4 (2026-06-10) · FGN