Supervised learning is the most common type of machine learning.
**The recipe:**
1. Collect labelled data (emails tagged spam/not-spam)
2. Train a model on this data
3. The model learns the mapping: input → label
4. Feed it new unlabelled data → it predicts the label
**Classic examples:**
- Email spam detection
- House price prediction
- Image classification (is this a cat or dog?)
- Fraud detection
The 'supervision' comes from the labels — you're teaching the model what the right answer is.