Classical models built from the math you already know
A cup of tea that has gone lukewarm is not obviously hot or obviously cold. If two piles are forming on the counter, one of steaming mugs and one of iced glasses, the fairest label for that lukewarm cup might be something like 70% hot pile and 30% cold pile. A Gaussian mixture model (GMM) gives clustering exactly that flexibility: instead of forcing every point into one bucket the way k-means does, each point gets a probability of belonging to every cluster at once.
🔒 This is a Pro lesson — the interactive figure, worked examples, quiz and practice open with Pro access.
▶ Gaussian Mixtures: Soft Clustering