Clustering: Predicting Behavior Through Data

Companies are predicting your next move by grouping you with strangers who act exactly like you.

It’s called clustering, a massive part of unsupervised machine learning. Think of it like

high school cliques but for data. The algorithm looks at a mess of information and finds hidden

sub-groups, whether that’s specific cancer types or shoppers who love discounts.

You have K-means, which forces data into specific teams based on distance, or hierarchical

clustering that builds a giant family tree. You might think your taste is unique,

but to an algorithm, you are just a coordinate in a very predictable box.