Is there a way (e.g. package) to perform an unsupervised classification using the ISODATA clustering algorithm in R?

2 Answers 2


There is a GitHub page on ISODATA clustering in R that may be worth a check.

enter image description here


I believe that closest you are going to come is the unsuperClass function in the RStoolbox package (http://bleutner.github.io/RStoolbox/rstbx-docu/unsuperClass.html). This function implements a k-means and not ISOCLUSER but, it is an unsupervised approach that is, honestly, preferable.

  • Thank you for your answer! Why is this approach preferable?
    – M514
    Dec 1, 2017 at 2:41
  • K-Means provides flexibility in defining class centers as well as evaluating the margin function (silhouettes) to find supported k (number of clusters). Dec 1, 2017 at 2:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.