Scribd
Upload
28 views3 pages

41 ML

The document outlines the process of hierarchical clustering using Single Link and Complete Link methods based on a given distance matrix for five points. Single Link clustering merges points based on minimum distances, while Complete Link clustering merges based on maximum distances, leading to different hierarchical structures. The final summary compares the two methods, highlighting their distinct approaches to merging clusters and the resulting dendrograms.

Uploaded by

mohdzaffar0561
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
28 views3 pages

41 ML

The document outlines the process of hierarchical clustering using Single Link and Complete Link methods based on a given distance matrix for five points. Single Link clustering merges points based on minimum distances, while Complete Link clustering merges based on maximum distances, leading to different hierarchical structures. The final summary compares the two methods, highlighting their distinct approaches to merging clusters and the resulting dendrograms.

Uploaded by

mohdzaffar0561
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Let's solve the hierarchical clustering problem using Single Link and Complete Link

clustering based on the given distance matrix.

Step 1: Understanding the Distance Matrix

The table provides the pairwise distances between five points (P1, P2, P3, P4, P5).

P1 P2 P3 P4 P5
P1 0.00 0.10 0.41 0.55 0.35
P2 0.10 0.00 0.64 0.47 0.98
P3 0.41 0.64 0.00 0.44 0.85
P4 0.55 0.47 0.44 0.00 0.76
P5 0.35 0.98 0.85 0.76 0.00

Step 2: Single Link Clustering (Minimum Distance)


In single link clustering, the distance between two clusters is the minimum distance
between points in those clusters.

Iteration Steps (Single Link)

1. Find the closest pair:


o The smallest distance is 0.10 (between P1 and P2). Merge {P1, P2}.
2. Update distances (Minimum method)
o Compute new distances from {P1, P2} to other points using:
d(C,X)=min (d(P1,X),d(P2,X))d(C, X) = \min(d(P1, X), d(P2, X))
o Updated distances:
o {P1, P2} - P3 → min(0.41, 0.64) = 0.41
o {P1, P2} - P4 → min(0.55, 0.47) = 0.47
o {P1, P2} - P5 → min(0.35, 0.98) = 0.35
3. Find next closest pair:
o The smallest distance is 0.35 ({P1, P2} and P5). Merge {P1, P2, P5}.
4. Update distances:
5. {P1, P2, P5} - P3 → min(0.41, 0.85) = 0.41
6. {P1, P2, P5} - P4 → min(0.47, 0.76) = 0.47
7. Find next closest pair:
o The smallest distance is 0.41 ({P1, P2, P5} and P3). Merge {P1, P2, P5, P3}.
8. Merge remaining clusters:
o Merge {P1, P2, P5, P3} and P4 at 0.44.

Dendrogram (Single Link)


P1 P2 P3 P4 P5
| | | | |
----(0.10)---- | |
| | |
------(0.35) |
| |
----(0.41)----
|
---(0.44)---

Step 3: Complete Link Clustering (Maximum Distance)


In complete link clustering, the distance between two clusters is the maximum distance
between points in those clusters.

Iteration Steps (Complete Link)

1. Find the closest pair:


o The smallest distance is 0.10 (P1 and P2). Merge {P1, P2}.
2. Update distances (Maximum method)
o Compute new distances from {P1, P2} to other points using:
d(C,X)=max (d(P1,X),d(P2,X))d(C, X) = \max(d(P1, X), d(P2, X))
o Updated distances:
o {P1, P2} - P3 → max(0.41, 0.64) = 0.64
o {P1, P2} - P4 → max(0.55, 0.47) = 0.55
o {P1, P2} - P5 → max(0.35, 0.98) = 0.98
3. Find next closest pair:
o The smallest distance is 0.47 ({P1, P2} and P4). Merge {P1, P2, P4}.
4. Update distances:
5. {P1, P2, P4} - P3 → max(0.64, 0.44) = 0.64
6. {P1, P2, P4} - P5 → max(0.98, 0.76) = 0.98
7. Find next closest pair:
o The smallest distance is 0.64 ({P1, P2, P4} and P3). Merge {P1, P2, P4, P3}.
8. Merge remaining clusters:
o Merge {P1, P2, P4, P3} and P5 at 0.98.

Dendrogram (Complete Link)


P1 P2 P3 P4 P5
| | | | |
----(0.10)---- | |
| | |
------(0.47) |
| |
----(0.64)----
|
---(0.98)---

Final Summary
Aspect Single Link Clustering Complete Link Clustering
Minimum distance between
Cluster Distance Maximum distance between clusters
clusters
Aspect Single Link Clustering Complete Link Clustering
First Merge P1 and P2 (0.10) P1 and P2 (0.10)
Next Closest Merge (P1, P2) and P5 (0.35) (P1, P2) and P4 (0.47)
Final Merge (P1, P2, P5, P3) and P4 (0.44) (P1, P2, P4, P3) and P5 (0.98)
Dendrogram Merges based on overall cluster
Builds from closest points first
Growth distances

Both methods form different hierarchical cluster structures, with single-link favoring chain-
like structures and complete-link forming more compact clusters.

Let me know if you need any modifications or explanations!

You might also like