k-Means Clustering

• We already learned in Module 1 that Clustering is about grouping similar things together, without labels.
• Now, let’s study one of the most popular clustering algorithms: K-Means.

1. What is K-Means?

• K-Means groups data into K clusters (K = number of groups you want).
• Each cluster has a center point (called a centroid).
• Every data point belongs to the cluster whose centroid it is closest to.

Example: Suppose you want to divide customers into 3 groups based on their income and age.

• K-Means will place 3 centers.
• Each customer will be assigned to the nearest center.
• The groups form automatically.

2. How Does K-Means Work?

Step by step intuition:

1. Choose K → Decide how many clusters (groups) you want.
2. Place centers → K-Means starts with K random points as “centers.”
3. Assign points → Each data point is assigned to the nearest center.
4. Update centers → Move each center to the average position of its assigned points.