AI Foundationsbeginner14 min · 10 steps
Unsupervised Learning: Clustering & Dimensionality Reduction
Find structure in data nobody labeled — group similar points with k-means, hierarchical clustering or DBSCAN, and squeeze many features into a few with PCA and t-SNE.
dataresponse
ProblemStep 1 of 10
Thousands of customers, zero labels
A shop has visit frequency and basket size for every customer, but nobody has tagged them as “bargain hunters” or “big weekly shoppers”. There is no y to learn from.
Unsupervised learning looks for structure in the inputs alone: groups of similar points, a few directions that explain most of the variation, or points that don't fit anywhere (anomalies).
InsightNo labels means no accuracy score. Results are judged by usefulness — do the groups make business sense, does the 2-D map reveal something?
← Previous lessonSupervised Learning: Regression & ClassificationNext lesson →Gradient Descent & Optimizers
Lesson 16 of 34 in AI Foundations