Navigation: <-- Part IV: Data Preparation | Main Index | Part VI: Principles That Transfer (Reflection) -->

---

Part V: Supervised Learning

Supervised learning is the family of ML methods where every training example comes with a known answer: a numeric value for regression tasks or a label for classification tasks. The model learns a mapping from inputs to outputs that generalizes beyond the training examples it saw. In this part you complete your first full inner-loop pass through the CRISP-DM Modeling and Evaluation phases - fitting real models to data, evaluating how well they generalize, and building the practical vocabulary every subsequent part assumes.

---

Nuggets in This Part

#	Nugget	Prerequisites
1	Supervised Learning	--
2	Linear Regression	Data Splits
3	Gradient Descent	Linear Regression
4	Underfitting and Overfitting	Linear Regression · Data Splits · Feature Engineering
5	Regularized Regression	Linear Regression · Gradient Descent
6	Hyperparameter Optimization	Regularized Regression · Underfitting and Overfitting
7	Classification Tasks	Supervised Learning
8	Classification Evaluation	Classification Tasks
9	Decision Trees	Supervised Learning · Classification Evaluation
10	Random Forests	Decision Trees
11	Logistic Regression	Linear Regression · Classification Tasks

---

Script v1.7 (2026-07-28) · FGN