"Text Mining - How to?"

dww

Hi I am very new to Rapid-I / Rapid Miner and was hoping someone might be able to help me get my head around some text mining issues.

I have a data set which is all free text responses to a question.  Each answer is on a separate line.  I am wanting to do some clustering/visualisation/analysis of the key themes that come out in the question responses.  I can successfully import my text (although it seems to make it all the heading of the columns with no actual data).

Can someone give me an idea of how I might format my data accordingly such that it imports fine for text mining.

Also any additional tips on how I might be able to use it for visualisation would be a great help.

Thanks in advance
DW

land

Hi,
the Text Mining Plugin provides some very usefull example processes. If you visit the download section of the text plugin on source forge, there is a file called rapidminer-text-4.3-examples.zip  which providing them. It might help if you take a look in the tutorial, too. It is available on the same download page.

Greetings,
  Sebastian