K-Means is a popular unsupervised learning algorithm used for clustering, it partitions a dataset into k (k is number of clusters needed) distinct, non-overlapping groups (clusters) based on similarity.

Working of K-means

• Consider a dataset with n data points and a desired number of clusters k:
• Initialize:
• Choose k cluster centroids randomly.
• Assignment Step:
• Assign each data point to the nearest centroid (based on Euclidean distance).
• Update Step:
• Recalculate the centroids as the mean of all points assigned to each cluster.
• Repeat:
• Steps 2 and 3 are repeated until:Centroids no longer move significantly (convergence), or A maximum number of iterations is reached.

Advantages

• Fast and efficient for large datasets.
• Easy to implement and interpret.
• Works well with spherical, well-separated clusters.

Limitations

• Must specify k beforehand.
• Sensitive to outliers and initial centroids.
• Assumes clusters are isotropic (uniform in all directions) and equally sized.
• Poor performance on non-convex clusters or clusters of different densities.