IAE Academy

GATE 2024 DA Q.42: solution and answer

GATE 2024 · DAQ.42 · 2 marks · MCQ

Consider the table below, where the element of the table is the distance between points and . Single linkage clustering is performed on data points, .

| | | | | | |
|---|---|---|---|---|---|
| | 0 | 1 | 4 | 3 | 6 |
| | 1 | 0 | 3 | 5 | 3 |
| | 4 | 3 | 0 | 2 | 5 |
| | 3 | 5 | 2 | 0 | 1 |
| | 6 | 3 | 5 | 1 | 0 |

Which ONE of the following is the correct representation of the clusters produced?

GATE 2024 DA question 42: A 5x5 symmetric distance table (given in the stem) and four dendrograms (A to D), each with leaves in the order x1, x2…Tap to open full size
Figure from the GATE 2024 DA paper. It is part of the question.
Figure description

A 5x5 symmetric distance table (given in the stem) and four dendrograms (A to D), each with leaves in the order x1, x2, x3, x4, x5.

  1. (A)

    Figure: dendrogram in which x1 and x2 join as a pair; x4 and x5 join as a pair, and x3 then joins {x4,x5}; the two groups join at the top.

  2. (B)

    Figure: dendrogram in which x1 and x2 join, then x3 joins {x1,x2}; x4 and x5 join as a separate pair; the two groups join at the top.

  3. (C)

    Figure: dendrogram in which x1 and x2 join; x3 and x4 join, then x5 joins {x3,x4}; the two groups join at the top.

  4. (D)

    Figure: dendrogram in which x1 and x2 join; x3 and x4 join; these two pairs merge; x5 joins last at the top.

Answer and solution

ANSWER

The correct dendrogram is option A, the tree .

Official answer: ASource: Official answer key, GATE 2024 DA (IISc Bengaluru)

  1. Single linkage merges the two clusters with the smallest minimum pairwise distance. The smallest entries in the table are and .

  2. Merge at height 1 and at height 1.

  3. Recompute the cluster distances with the minimum over pairs: , , .

  4. The smallest is 2, so joins at height 2, giving .

  5. Finally over all cross pairs , so the two groups merge at height 3.

  6. The tree is : a pair , a pair with joining it, and the two groups joined at the top. This is option A.

  7. Option B has joining while stay a separate pair, but is closer to (distance 2) than to (distance 3). Option C pairs with and then adds , but and (distance 1) merge before can join either. Option D merges as a pair and leaves until last, but is closest to (distance 1).

KEY IDEA

In single linkage, the distance between two clusters is the smallest pairwise distance across them. Merge the closest pair repeatedly, recomputing cluster distances with a minimum.

WATCH OUT FOR

Merging with as a pair, because looks small. That ignores that and already merged at distance 1, so joins the cluster rather than alone.

From the GATE 2024 DA paper. Every question of this paper · Download the full paper · All 4 Clustering questions, with notes

More Clustering questions