Clustering
Last Updated: July 29, 2026 | By Mihail Sebastian | AI Dictionary
An unsupervised learning technique that groups similar data points so structure emerges from data that carries no labels or predefined categories.
What is Clustering?
Clustering is an unsupervised learning technique that groups similar data points together, without labels or predefined categories to guide it. The groups emerge from the data itself: the algorithm measures similarity between points and pulls the alike ones into clusters.
That is the practical difference from classification: classification learns categories from data where the right answer is given; clustering has to discover the categories on its own.
Types of Clustering
- K-means clustering splits data into a chosen number of clusters (k) and shuffles points between them until each cluster is as tight as possible.
- Hierarchical clustering builds a tree of clusters within clusters, so you pick the level of granularity after the fact.
- DBSCAN groups points by density, which lets it find clusters of odd shapes and mark stray points as noise.
Applications of Clustering
- Customer segmentation: groups customers by behavior so marketing addresses real segments rather than guessed ones.
- Image segmentation: splits an image into meaningful regions, used in medical imaging and object detection.
- Anomaly detection: points that fit no cluster are candidates for fraud or equipment faults.
Example of Clustering
An online retailer feeds two years of purchase histories into a k-means model: order frequency, basket size, discount usage, product categories. The algorithm returns four clusters nobody specified in advance. One turns out to be weekly regulars with small baskets, another holiday-only shoppers with large ones, a third bargain hunters who buy only on discount, a fourth new customers still deciding.
The algorithm only draws the boundaries. Naming the segments, judging whether they mean anything, and deciding what to do about them stays with the humans reading the results.
Related AI terms: Classification · Anomaly Detection · K-Means Clustering · Unsupervised Learning · Dimensionality Reduction
Did you like the Clustering gist?
Learn about 250+ need-to-know artificial intelligence terms in the AI Dictionary.
Mihail Sebastian — Writes about AI governance, regulation, and the technology behind them. Placeholder bio — replace with a real credential line. About