Also known as: K means Clustering
Кластеризация k-средних — метод обучения без учителя, который группирует примеры в заданное число кластеров на основе близости к центроидам кластеров. Она широко используется для сегментации, обнаружения паттернов и разведочного анализа, но требует заранее выбрать число кластеров.
A method of vector quantization, originally from signal processing, that aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean (cluster centers or cluster centroid), serving as a prototype of the cluster.
A side-by-side comparison of K-means Clustering and DBSCAN. Understand how centroid-based grouping differs from density-based clustering with noise detection.
A side-by-side comparison of K-means Clustering and Hierarchical Clustering. Understand how flat centroid-based grouping differs from tree-like nested cluster structures.