This notebook explores a way to analyze different kmeans models analyticaly, using accuracy for clustering, instead of plots. It then shows the best and worst models created by different values of number of clusters and attributes used for clustering.
The repository has a Jupyter Notebook, a pdf and an html file all with the same content