Centroid-based clustering is a method that organizes data into non-hierarchical clusters. The process relies on identifying the centroid of a cluster, which is the arithmetic mean of all data points contained within that cluster.
Algorithms like K-Means fall into this category. While centroid-based algorithms are considered efficient, they are generally sensitive to initial conditions and the presence of outliers in the data. This approach is utilized in marketing for customer segmentation based on numeric data like purchase frequency and for clustering product images.