how to use kmeans?

it_mjjiang
it_mjjiang New Altair Community Member
edited November 5 in Community Q&A
new to RapidMiner. How to use kmeans: i.e., how to prepare data set (what is the format), how to tune parameters, how to use generated cluster?

Please gives me a link to its documentation if there is.

Thanks!
Tagged:

Answers

  • Hello

    The issue with k-means is selecting the value of k and there is no right answer. You have to use various validity measures to help direct your search for the best clustering. It may also be the case that your data does not lend itself to an efficient clustering becaue the clusters are not spherical and of similar density. In this case, you would have to use other clustering algorithms.

    I made an example http://rapidminernotes.blogspot.com/2011/03/counting-clusters-part-ii.html

    regards

    Andrew
  • IngoRM
    IngoRM New Altair Community Member
    Hi,

    well, you have noticed the complete set of documentation around RapidMiner already, didn't you? Beside our multi-hundreds-of-thousands-dollars-license-costs-software-for-free we also offer Not enough? Well, time is money. Get in contact with Rapid-I, ask for an offer for our Support Subscriptions. Or our training courses. Or our webinars. Or...

    I am sure that the documents above will explain everything necessary for your specific problem. The only thing you need is time to work through it  ;D

    Cheers,
    Ingo
  • In my experience, examples from others like the ones mentioned by Ingo are the key and that's how I learnt most of what I know about RapidMiner. Now, when I create a new process, I try to give something back to the ecosystem for others to learn from.

    Andrew