The K-means algorithm is a famous unsupervised learning algorithm used for data analysis and clustering. It divides the dataset into K clusters, with each cluster having its own centroid. The algorithm assigns each point to the closest cluster by calculating distances