Question 26
UnclassifiedK-Means clustering requires the user to specify:
Correct answer: B
Explanation
K-Means is a partitioning algorithm that groups data into a fixed number of clusters, so the value of K must be set before training begins. The algorithm then assigns points to the nearest centroid and iterates until the clusters stabilize.
Why each option is right or wrong
A. The labels of training samples
B. The number of clusters K in advance
Under the standard K-means procedure, the algorithm cannot begin without a preset value for the number of centroids, because the objective function is defined over exactly K partitions and K initial cluster centers. In practice, the user must choose K before fitting the model; the iterative assignment/update steps then proceed until convergence, but K itself is not learned from the data during training.
C. The kernel function
D. The depth of trees