The k in k-Means indicates ________.

• A. The Distance Metric
• B. The Number Of Clusters
• C. The Kernel Type
• D. The Initialization Method

Answer: (B)

In k-means, the k represents the number of clusters you want the data to be partitioned into. The algorithm creates k centroids, assigns each data point to the nearest centroid, and then updates those centroids iteratively until the assignments stabilize. The value of k directly determines how many groups you will end up with; if you set k to a different number, the resulting partitioning changes accordingly.

Distance metric and initialization method are separate aspects. The distance metric (often Euclidean) is how you measure closeness to a centroid, but it doesn’t define how many clusters there are. Initialization method affects where the starting centroids begin, which can influence the final result, but again doesn’t determine the number of clusters. Kernel type isn’t part of standard k-means; traditional k-means uses a centroid-based approach, not a kernel-based method. For example, choosing k = 3 means you’re aiming for three distinct groups in the data.