When comparing k-means clustering and hierarchical clustering for a…

Question

When comparing k-means clustering and hierarchical clustering for a dataset with 500,000 observations in SAS, which statement is most accurate?

Accepted Answer

K-means scales better to large datasets because it is O(n) per iteration, while hierarchical clustering is O(n^2) to O(n^3) — Here's the deal — b is correct because agglomerative hierarchical clustering requires computing and storing a pairwise distance matrix (O(n^2) space) and O(n^2) to O(n^3) time, making it infeasible for 500,000 observations. K-means iteratively reassigns points to the nearest centroid, which is O(nkiterations) and scales far better.

Answer

Hierarchical clustering is faster than k-means on large datasets because it requires only one pass through the data

Answer

Hierarchical clustering always produces better cluster quality because it uses the complete linkage criterion

Answer

K-means and hierarchical clustering have identical computational complexity for datasets over 100,000 rows

When comparing k-means clustering and hierarchical clustering for a dataset with 500,000 observations in SAS, which statement is most accurate?

Short Explanation and Infographic

Full Explanation