KEMBAR78
ML K-Means | PDF
Scribd
Upload
12 views3 pages

ML K-Means

The document illustrates a K-Means clustering example with four data points, aiming to group them into two clusters using the K-Means algorithm. Initial centroids are chosen, distances to these centroids are calculated, and points are assigned to clusters based on proximity. After recalculating centroids and reassessing point assignments, the final clusters remain unchanged, indicating the algorithm has converged.

Uploaded by

mohammadsameerfnadaf
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
12 views3 pages

ML K-Means

The document illustrates a K-Means clustering example with four data points, aiming to group them into two clusters using the K-Means algorithm. Initial centroids are chosen, distances to these centroids are calculated, and points are assigned to clusters based on proximity. After recalculating centroids and reassessing point assignments, the final clusters remain unchanged, indicating the algorithm has converged.

Uploaded by

mohammadsameerfnadaf
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

K-Means Clustering Example:

We are given 4 data points: A(2,3), B(3,3), C(8,6), and D(9,5). We want to cluster
them into 2 groups using the K-Means algorithm with K=2. We'll use Euclidean
distance to determine the closeness of points to centroids.

Step 1: Choose Initial Centroids

Let’s pick A(2,3) and C(8,6) as the initial centroids:


- Centroid 1 (Cluster 1): A(2,3)
- Centroid 2 (Cluster 2): C(8,6)

Step 2: Assign Points to the Nearest Centroid (Using Euclidean Distance)

Euclidean Distance Formula:


d = √[(x₂ − x₁)² + (y₂ − y₁)²]

Calculations:
- A(2,3) to A = 0
- A(2,3) to C = √[(8−2)² + (6−3)²] = √45 ≈ 6.71
- B(3,3) to A = √1 = 1.0
- B(3,3) to C = √34 ≈ 5.83
- C(8,6) to A = √45 ≈ 6.71
- C(8,6) to C = 0
- D(9,5) to A = √53 ≈ 7.28
- D(9,5) to C = √2 ≈ 1.41
Point Distance to Distance to Assigned Cluster
A(2,3) C(8,6)

A(2,3) 0 6.71 Cluster 1

B(3,3) 1.0 5.83 Cluster 1

C(8,6) 6.71 0 Cluster 2

D(9,5) 7.28 1.41 Cluster 2

Resulting Clusters:
- Cluster 1: A(2,3), B(3,3)
- Cluster 2: C(8,6), D(9,5)

Step 3: Recalculate Centroids

Compute the new centroids by taking the average of points in each cluster.

New Centroid 1 (A and B):


x = (2 + 3)/2 = 2.5
y = (3 + 3)/2 = 3.0 → (2.5, 3)

New Centroid 2 (C and D):


x = (8 + 9)/2 = 8.5
y = (6 + 5)/2 = 5.5 → (8.5, 5.5)

Step 4: Reassign Points Based on New Centroids

Recalculate distances to new centroids:


- A to (2.5, 3) = 0.5
- B to (2.5, 3) = 0.5
- C to (8.5, 5.5) ≈ 0.71
- D to (8.5, 5.5) ≈ 0.71
Point Distance to (2.5, Distance to (8.5, Assigned Cluster
3) 5.5)

A(2,3) 0.5 6.96 Cluster 1

B(3,3) 0.5 6.04 Cluster 1

C(8,6) 6.26 0.71 Cluster 2

D(9,5) 6.8 0.71 Cluster 2

Final Clusters
Cluster 1: A(2,3), B(3,3)
Cluster 2: C(8,6), D(9,5)

As the cluster assignments didn’t change after updating the centroids, the algorithm
stops here.

You might also like