"Beginner's question: Clustering using only a few attributes"

unbeliever

Hello, I was wondering wether it is possible to run a clustering algorithm in RapidMiner using only a selection of the available attributes per case, so I can play around a little with them and find the best set for my purposes. Also having a String in a case helps to identify the case, but I don't want it to be used in the clustering process.

thx in advance

TobiasMalbrecht

Hi,

sure you can do that. You can filter the attributes you do not wish to include into the analysis from the attribute set by applying an [tt]AttributeFilter[/tt] operator with the appropriate parameter settings.

Regards,
Tobias