Which of the following best describes the K-Means clustering algorithm?

K-Means is an unsupervised clustering algorithm that works on unlabelled data and discovers natural groupings. The algorithm works in four steps: Step 1 — Initialisation: Randomly select K points as initial centroids (K is defined by the user). Step 2 — Assignment: Each data point is assigned to the nearest centroid using Euclidean distance (or Manhattan distance). Step 3 — Update: Recalculate each centroid as the mean of all data points assigned to it. Step 4 — Repeat Steps 2 and 3 repeat until centroids no longer move significantly (convergence). The value of K must be chosen in advance with the Elbow Method (plotting Within-Cluster Sum of Squares / WCSS against K values and finding the "elbow" point). This helps determine the optimal K. K-Means assumes clusters are spherical and roughly equal in size, is sensitive to outliers, and is sensitive to initial centroid placement (solved by K-Means++ initialisation). Banking applications include Customer segmentation (grouping customers by transaction behaviour — high-value, medium, dormant), Branch performance clustering (grouping branches with similar deposit/advance patterns), Fraud detection (identifying unusual transaction clusters that deviate from normal spending groups) and ATM usage pattern clustering (optimising cash replenishment by grouping ATMs with similar withdrawal patterns).