Classical models built from the math you already know
Every classifier built so far in this course has drawn one boundary and asked which side of it a point falls on. Logistic regression, the decision-boundary picture, the support vector machine: all of it assumed exactly two classes. Most real classification problems have more than two. A handwritten digit is one of ten. A support ticket is one of a dozen categories. A flower is one of several species. None of that earlier machinery gets thrown away. It just needs a plan for turning several two-class answers into one many-class answer.
🔒 This is a Pro lesson — the interactive figure, worked examples, quiz and practice open with Pro access.
▶ Multiclass, the Confusion Matrix & Metrics