k-means operator in RapidMiner vs simplekmeans in weka

prat87

Hello,

I am clustering my dataset using k-means on 2 tools:
weka
rapid miner (k-means operator)

The results are very different. And the results form rapidMIner is infact the one I was looking for.

It will be very very helpful if anybody can answer my questions below:
1) What is different in weka's simplekmeans and rapidminer's k-means operator ?
2) Is there a way to know all the paramters used in rapidminer's k-means operator - so that I can replicate the same result on weka or any other tool ?
3) Is Max_runs  (max number of runs of k-Means with random initialization that are performed) = seed in weka's simpleKMeans?

Thank you. I really appreciate your time.

raazzaa

Hello Prat87,

I've been having the same problems as yours, have you found any solutions yet?
It'd be great if you share me what you've found since you got that problem couple months ago.

thank you