Solution

ID: past-exam-of-the-mathematics-course-of-the-university-of-cambridge/2007/iii/paper-47/4/c/solution

In agglomerative hierarchical clustering, start with eight singleton groups. Single-linkage clustering uses the smallest distance between cross-pairs; complete-linkage clustering uses the largest. Updating these cluster distances after each merge, one valid order of the tied merges gives
The merges at height may be reordered; they give the same requested three-cluster cuts. For example, the single-link distance from to is , while its complete-link distance is their maximum . After forming , its single-link distance to is . The complete-link distance between and is . Finally the closest cross-pair between and is at , while the farthest is at .
At three groups the partitions are therefore
Two groups, and , are more strongly supported by the merge heights. For single linkage the three-to-two merge is at , close to the preceding height , whereas the two-to-one merge is delayed until . For complete linkage the three-to-two merge at is close to , whereas the final merge jumps to . Both methods agree on the two-group partition but split its second group differently when forced to return three groups. A three-group choice is possible if required substantively, but is not particularly compelling from these dissimilarities alone.

New to topics? Read the docs here!