8.6 Clustering
Clustering groups similar data points together without being told the groups in advance.
Chapter 8: Data and AI Algorithms
Clustering is the most common unsupervised learning task. The algorithm looks at the data and puts similar points into groups (clusters).
A popular algorithm, K-means, works like this: pick K starting centre points, assign each data point to the nearest centre, move each centre to the middle of its group, and repeat until things stop changing.
It's used for customer segments, grouping similar news stories, organising photos and spotting unusual behaviour.
Key takeaways
Clustering = unsupervised grouping
K-means: assign, re-centre, repeat
Used for segments, news grouping, anomaly spotting