Stopping Criteria for K-means

The K-Means algorithm does not run forever. It stops when any one of the following conditions is met:

1. Centroids Do Not Change

  • If the centroids remain the same even after multiple iterations,
  • It means the algorithm has converged.

2. Points Remain in the Same Cluster

  • If no data point changes its cluster membership,
  • The clusters are considered stable.

3. Maximum Number of Iterations Is Reached

  • A maximum limit (e.g., 100 iterations) is set in advance.
  • Once this limit is reached, the algorithm stops automatically.