how to use kmeans?

it_mjjiang

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!

awchisholm

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

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

• An English and a German Manual of about 150 pages: http://rapid-i.com/content/view/36/210/
• More than 50 videos from us and many more from community members: http://rapid-i.com/content/view/189/212/
• A developer manual for extending RapidMiner in our shop: http://rapid-i.com/component/page,shop.product_details/flypage,flypage.tpl/product_id,52/category_id,5/option,com_virtuemart/Itemid,180/
• A second (planned) one which will be released shortly (look here in the forum for more details)
• A set of samples directly delivered with RapidMiner
• A group on myExperiment.org and more than 100 processes showing nice tricks with RapidMiner on this platform: http://www.myexperiment.org
• This forum here with many helpful people willing to help you - if they know your problem in detail.
• A Wiki containing a description of all operators - and which could be improved - also by you!
• Hey, and finally it's open source - of course you also have access to the source code and the API

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

awchisholm

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