K-Means Clustering DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 4)

Questions 16–20

1. 16foundation · easy

   Which distance metric is the default and most commonly used in standard K-means clustering?

   • ACosine similarity
   • BEuclidean distance
   • CHamming distance
   • DManhattan distance
   Reveal Solution Discussion Select an answer first
2. 17application · medium

   A data scientist is explaining the K-means algorithm to a colleague. Which sequence correctly describes the iterative process?

   • AAssign points to the nearest centroid, initialize centroids, update centroids, repeat until convergence.
   • BInitialize centroids, update centroids to the mean of all points, assign points to the nearest centroid, repeat until convergence.
   • CInitialize centroids, assign each point to the nearest centroid, update centroids to the mean of assigned points, repeat until convergence.
   • DInitialize centroids, assign points to the nearest centroid, update centroids to the median of assigned points, repeat until convergence.
   Reveal Solution Discussion Select an answer first
4. 18application · medium

   A data scientist is running K-means on a large dataset and notices that the centroids are still moving noticeably after 100 iterations. The scientist wants to ensure the algorithm stops when the centroids have essentially stabilized, but also wants to avoid excessive computation. Which stopping criterion is most appropriate?

   • AStop after a fixed number of iterations, such as 100, regardless of centroid movement.
   • BStop when the assignment of points to clusters changes by less than 1% between iterations.
   • CStop when the change in centroid positions between iterations falls below a small threshold, with a maximum iteration limit as a safety net.
   • DStop when the total within-cluster sum of squares (WCSS) stops decreasing by more than 0.1%.
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   A data scientist runs K-means with k values from 2 to 10 and plots the total within-cluster sum of squares (WCSS) against k. The plot shows a sharp decrease from k=2 to k=4, then a gradual decrease from k=5 onward. Which k is most appropriate to choose?

   • Ak=2
   • Bk=5
   • Ck=4
   • Dk=10
   Reveal Solution Discussion Select an answer first
6. 20foundation · easy

   What is a common stopping criterion for K-means besides centroid movement?

   • AA minimum cluster size.
   • BA maximum number of iterations.
   • CA minimum number of clusters.
   • DA maximum distance between points.
   Reveal Solution Discussion Select an answer first