Due to recent updates, all users are required to create an Altair One account to login to the RapidMiner community. Click the Register button to create your account using the same email that you have previously used to login to the RapidMiner community. This will ensure that any previously created content will be synced to your Altair One account. Once you login, you will be asked to provide a username that identifies you to other Community users. Email us at Community with questions.

"How to calculate Silhouette validation for k-means using R in Rapidminer?"

xixirhwfyxixirhwfy Member Posts: 9 Contributor II
edited June 2019 in Help
Hello everyone,
  I'm trying to use Silhouette validation for k-means clustering, but there's no such operator. So I'm trying to add R "execute script" operator to calculate silhouette. But I don't know much about R language. I want to calculate k-means clustering using operator in Rapidminer and only do the silhouette analysis in the R script. Does anyone know how to write such script? And how to connect the Rapidminer operator together with the R operator to make the whole process work and how the parameter keys are set? Better to make an example using XML, thank you!
Tagged:

Answers

  • awchisholmawchisholm RapidMiner Certified Expert, Member Posts: 458 Unicorn
    Hello,

    How about this?

    http://rapidminernotes.blogspot.co.uk/2011/06/counting-clusters-part-r.html

    Andrew
  • xixirhwfyxixirhwfy Member Posts: 9 Contributor II
    Thank you, Andrew! But I want to use kmeans clustering not the k-medoid(pam()?), and I want to use the operator in Rapidminer to calculate kmeans, then connect to a "R:execute script" operator which only calculate the Silhouette value, could you please make another simpler example to show this? Because I know nothing about the R language. Thank you very much!
  • awchisholmawchisholm RapidMiner Certified Expert, Member Posts: 458 Unicorn
    Hello,

    As it happens, I'm contributing a chapter to an upcoming book - http://rapidminerbook.com - that does something even closer to what you want. I'm afraid you'll have to wait for that. You could have a go yourself of course :-)

    Andrew
Sign In or Register to comment.