Question

Which problem-solving approach does the K-means algorithm employ?

• A. Gradient Descent
• B. Expectation-Maximisation
• C. Decision Trees
• D. Support Vector Machines

Hint:

K-means: Assign clusters (Expectation) → Update centers (Maximisation).

Answer 1

The Correct Option is B

K-means is an unsupervised clustering method that partitions data into clusters by minimizing variance within each cluster. It employs an iterative process analogous to the Expectation-Maximisation (EM) algorithm. Initially, data points are allocated to clusters (the Expectation phase), followed by the refinement of cluster centroids (the Maximisation phase). This cycle continues until cluster assignments remain constant. Gradient Descent is an optimization technique for continuous functions and is not directly applied to clustering. Decision Trees and Support Vector Machines are distinct supervised learning algorithms used for classification and regression tasks. Therefore, K-means operates on an Expectation-Maximisation-based methodology.