Question 40
Domain 2 — Data, Machine Learning, and Model DevelopmentWhich of the following methods scales poorly with large data sets?
Correct answer: C
Explanation
Hierarchical clustering scales poorly because it must repeatedly merge or split clusters while tracking relationships among many points, which becomes computationally expensive as data grows. In practice, its time and memory demands increase much faster than methods designed for large data sets.
Why each option is right or wrong
A. Fuzzy
B. Density-based
C. Hierarchical
Hierarchical clustering has poor scalability because the standard agglomerative procedure recomputes inter-cluster distances at each merge step, with a typical time complexity of about O(n^2) to O(n^3) and memory use that can also grow on the order of O(n^2). For large n, the repeated pairwise comparisons and dendrogram construction become computationally expensive compared with partition-based methods such as k-means.