K-Means Clustering Explained (with Example)

What is K-Means Clustering?

K-Means Clustering is an unsupervised learning algorithm that groups data into K clusters based on similarity.

• K = number of groups you want
• Means = average (center of each group)

---

Intuition

You tell the algorithm:

“Divide this data into K groups.”

1. Place K random centers (centroids)
2. Assign each point to nearest center
3. Recalculate the center
4. Repeat until stable

---

Step-by-Step

1. Choose number of clusters (K)
2. Initialize K centroids randomly
3. Assign each data point to nearest centroid
4. Update centroids (mean of points)
5. Repeat until no change

---

Example

Student Marks:

[35, 40, 45, 70, 75, 80]

K = 2

Initial Centroids:

C1 = 40
C2 = 75

Cluster Assignment:

• Cluster 1 → 35, 40, 45
• Cluster 2 → 70, 75, 80

Updated Centroids:

C1 = 40
C2 = 75

Algorithm stops as no change occurs.

---

Final Clusters

• Cluster 1 → Low scores
• Cluster 2 → High scores

---

Real-Life Use Cases

• Customer segmentation
• Image compression
• Document grouping
• Market analysis

---

Advantages

• Simple and fast
• Works well for large datasets

Limitations

• Need to choose K beforehand
• Sensitive to initial centroids
• Not good for irregular shapes

---

Summary

K-Means groups similar data into K clusters by updating cluster centers iteratively.