K-Means Clustering Demo
Step-by-step visualization of the k-means clustering algorithm
Configuration
Number of Clusters (k):
3
Number of Data Points:
Generate Data
Initialize Centroids
Step
Auto Run
Reset Centroids
Clustering Visualization
Download Data
Cluster Statistics
Convergence Progress
Elbow Method (Optimal k)
Help
K-Means Algorithm
Step 1:
Initialize k random centroids
Step 2:
Assign each point to nearest centroid
Step 3:
Update centroids to mean of assigned points
Step 4:
Repeat steps 2-3 until convergence
Understanding Metrics
WCSS:
Within-Cluster Sum of Squares - lower is better
Elbow Method:
Plot WCSS vs k to find optimal number of clusters
Convergence:
Algorithm stops when centroids don't move
Local Optima:
Different initializations may give different results