The averages for each attribute in each cluster created by a k-Means model are called _______.

• A. Means
• B. Coordinates
• C. Medians
• D. Centroids

Answer: (D)

In k-means, the center of each cluster is defined as the centroid, which is the average position of all points in that cluster across every attribute. This centroid is a vector whose components are the mean values for each feature among the points assigned to that cluster. It serves as the representative center used to decide which cluster a new point belongs to and to re-center the clusters during the iterative process. The term centroids captures both the idea of being a center and being calculated from the means of the cluster’s members. While the individual per-feature averages are indeed means, the standard name for the cluster’s center is centroid. Medians would pertain to a different method (k-medians), and coordinates describe the location rather than naming the center itself.