The k-Means clustering technique for data analysis is ideal for _________.

• A. Classification
• B. Segmentation
• C. Regression
• D. Clustering

Answer: (B)

The main idea here is forming meaningful groupings of data based on similarity, which is segmentation. k-Means partitions the data into K clusters by assigning each observation to the nearest cluster center and updating those centers to minimize the within-cluster variation. This yields natural, homogeneous groups such as customer segments or market segments that you can analyze or tailor strategies around. It’s unsupervised, meaning it doesn’t require labeled data. The other goals don’t fit as well: classification relies on labeled examples to predict categories, and regression predicts a numeric value. Although clustering is the broader method, its practical use in many applications is to create segments, making segmentation the best fit.