Countries Clustering

We analyze here two clustering algorithms: the K-Means and the Gaussian Mixtures. The first aims to evenly distribute all datapoints in K clusters. The second aims to find the best K Gaussian distributions, each with its own probability, mean and variance.

We observe both models and differences, while following their evolution, in animated graphs at each step.

Our studies are applied to real data from countries around the world and we investigate their outcomes, meanings and draw a parallel with real-world features.