Machine learning splits into two great families, depending on whether the examples come with answers attached.

• Supervised learning. Every example carries a correct label — a photo tagged "cat", a house with its sale price. The model learns to predict the label from the features. This covers regression (predict a number) and classification (predict a category).
• Unsupervised learning. No labels at all — just raw data. The model hunts for structure on its own: natural clusters, or a simpler way to describe the data.

Labels, or no labels

The same points, two situations. In the supervised view they arrive pre-coloured by their true class, and the model learns the boundary between them. In the unsupervised view they're all one colour, and the model must discover the two groups itself. Toggle between them.

Which problem are you solving?

The question "do I have labelled answers?" decides everything downstream — which algorithms apply, how you measure success, even how much data you need. Most of this course is supervised (it's where the clearest ideas live), starting with linear regression; we return to the unsupervised world at the end. (There's a third cousin, reinforcement learning, where an agent learns from rewards — a story for another day.)