k-Means Clustering is a type of Hard Clustering that aims to partition 𝑛 observations into 𝑘 clusters

We note 𝑐^(𝑖) the cluster of data point 𝑖 and 𝜇_𝑗 the center of cluster 𝑗

Algorithm ― After randomly initializing the cluster centroids 𝜇₁, 𝜇₂, …, 𝜇_𝑘 ∈ ℝ^𝑛, the k-means clustering repeats the following step until convergence:

K-Means - General Algorithm

iterative-clustering-algorithm(points, k) {
	cluster-centers = k random points (means)
	do until convergence:
		for each point in points:
		assign point to closest cluster-center
		change each cluster-center to the average of its assigned points
}

K-Means - Other

• Louvain Method vs K-Means
• K-Means Clustering vs Fuzzy C-Means Clustering