Introduction to K-Means Clustering Answers 1. What is a primary limitation of the K-means algorithm? A. Focuses only on categorical data B. Requires iterative adjustments of features C. Assumes clusters are roughly spherical D. Automatically determines the number of clusters Answer: Assumes clusters are roughly spherical (C) K-means assumes that clusters are spherical in shape, which may not always apply to real-world datasets. 2. Which application is NOT commonly associated with K-means clustering? A. Image segmentation B. Customer segmentation C. Time-series forecasting D. Document clustering Answer: Time-series forecasting (C) K-means is typically not used for time-series forecasting, but is known for applications such as image segmentation, customer segmentation, and document clustering. 3. What is the purpose of K-means++? A. To reduce the number of clusters needed B. To eliminate outliers from the dataset C. To improve the initial centroid selection D. To enhance the performance of spherical cluster assumption Answer: To improve the initial centroid selection (C) K-means++ improves the selection of initial centroids to avoid local optima during clustering. 4. How does K-means handle outliers in the data? A. It identifies and removes outliers before clustering B. It allows outliers to skew the centroid locations C. It treats outliers as separate clusters automatically D. It gives equal weight to all data points regardless of their value Answer: It allows outliers to skew the centroid locations (B) K-means is sensitive to outliers as they can significantly affect the centroids and the overall clustering results. 5. What characteristic of K-means makes it challenging to apply in datasets with irregular shapes? A. The assumption of spherical clusters B. The use of a data subset for each iteration C. The requirement to preprocess the data extensively D. The need for standardization of features Answer: The assumption of spherical clusters (A) K-means struggles with irregularly shaped clusters because it assumes that the clusters are roughly spherical.