Supervised learning algorithms

Supervised learning algorithms learn a mapping from inputs X to outputs y using labeled training data. Each training example includes both features and the correct target label. The goal is to generalize this mapping to make accurate predictions on unseen data.

Common algorithm families:

• Linear models (e.g., linear/logistic regression) – fast, interpretable baselines.
• Tree-based methods (decision trees, random forests, gradient boosting) – handle mixed feature types, capture non‑linearities, often strong out‑of‑the‑box.
• Support Vector Machines (SVMs) – effective in high‑dimensional spaces, with kernel tricks for non‑linear boundaries.
• k‑Nearest Neighbors (k‑NN) – instance‑based, simple, relies on a distance metric.
• Neural networks – very flexible function approximators, scale to large and complex datasets.

Key tasks:

• Classification – predict discrete labels (spam vs. not spam).
• Regression – predict continuous values (house prices).

Performance is evaluated using a held‑out validation/test set with metrics such as accuracy, precision/recall/F1 (classification) or MSE/MAE/R² (regression).