Interesting the author neglects to mention the Gap statistic[1] for the stopping criteria. It's a bit of a mouthful, but it does provide a metric for determinin

Interesting the author neglects to mention the Gap statistic[1] for the stopping criteria. It's a bit of a mouthful, but it does provide a metric for determining how many clusters the data indicates are present as compared to a random distribution in the same dimensions. [1] https://academic.oup.com/jrsssb/article-abstract/63/2/411/7083348?redirectedFrom=fulltext https://academic.oup.com/jrsssb/article-abstract/63/2/411/70...