Understanding concept of Gaussian Mixture Models

Question

I\'m trying to understand GMM by reading the sources available online. I have achieved clustering using K-Means and was seeing how GMM would compare to K-means.

Here

Answer 1

Covariance tells you how the data varies in the space, if a distribution has large covariance, that means data is more spread and vice versa. When you have the PDF of a gaussian distribution (mean and covariance params), you can check the membership confidence of a test point under that distribution.

However GMM also suffers from the weakness of K-Means, that you have to pick the parameter K which is the number of clusters. This requires a good understanding of your data's multimodality.