Drawbacks of k-means Clustering you will notice that all the clusters created are circular. This is because the centroids of the clusters are updated iteratively using the mean value. Now, consider the following example where the distribution of points is not circular. What do you think will happen if we use k-means clustering on this data? It would still attempt to group the data points circularly. That’s not great! k-means fails to identify the right clusters: Hence, we need a different way to assign clusters to the data points. So instead of using a distance-based model, we will now use a distribution-based model. And that is where Gaussian Mixture Models come into this article! Introduction to Gaussian Mixture Models (GMMs) The Gaussian Mixture Model (GMM) is a probabilistic model used for clustering and density estimation. It assumes that the data is generated from a mixture of several Gaussian components, each representing a distinct clus...