Machine Learning Basics

What you'll learn

• Explain what machine learning is, the main problem types, and when ML beats hand-written rules.
• Prepare data correctly: features and labels, train/test splits, scaling and encoding, and avoiding leakage.
• Train and evaluate regression and classification models with the right metrics and a baseline.
• Recognise overfitting and underfitting and use cross-validation and regularisation to control them.
• Use ensembles, hyperparameter search and permutation importance, and apply clustering, PCA and anomaly detection.
• Run a small end-to-end project responsibly, including fairness and monitoring considerations.

Syllabus

What Machine Learning Is

1. Rules From Data Instead of Rules by Hand
2. Supervised, Unsupervised and Reinforcement Learning
3. The Machine Learning Workflow

Preparing Data

1. Features, Labels and Train/Test Splits
2. Data Leakage
3. Scaling and Encoding Features

Regression: Predicting Numbers

1. Linear Regression
2. Regression Metrics: MAE, RMSE and R²
3. Always Beat a Baseline

Classification: Predicting Categories

1. Logistic Regression and Probabilities
2. Decision Trees and k-Nearest Neighbours
3. Confusion Matrix, Precision and Recall
4. Choosing the Decision Threshold

Generalisation: Overfitting and Validation

1. Underfitting and Overfitting
2. Cross-Validation
3. Regularisation

Ensembles, Tuning and Interpretation

1. Random Forests and Gradient Boosting
2. Hyperparameter Search
3. Which Features Matter? Permutation Importance

Unsupervised Learning

1. Clustering With k-Means
2. Dimensionality Reduction With PCA
3. Anomaly Detection

Putting It Together Responsibly

1. An End-to-End Mini Project
2. Responsible Machine Learning
3. A Machine Learning Project Checklist