centroid clustering

Centroid clustering, primarily K-means, is an unsupervised machine learning technique: it partitions a dataset into 'K' clusters by iteratively assigning data points to the cluster with the nearest arithmetic mean (centroid).

This is a core partitioning method: we group data points by minimizing the sum of squared distances between each point and its cluster's centroid. The process requires you to specify 'K' (the number of clusters) upfront. The algorithm then follows an iterative loop: first, it assigns all data points to the nearest centroid (the cluster's mean); second, it recalculates the centroid based on the newly assigned points. This assignment-recalculation cycle repeats until the centroids stabilize (convergence). For example, in e-commerce, a centroid-based model can segment 100,000 customers into five distinct groups (K=5) based on purchase frequency and average transaction value, enabling targeted marketing strategies.

https://neptune.ai/blog/k-means-clustering-explained