"word count in different types of documents"

barthos
barthos New Altair Community Member
edited November 5 in Altair RapidMiner
Hi,

I would like to do something rather simple but I can't manage to do it.
I have an excel file with two columns each containing about 1000 rows. The first columns define the categorie (A,B,C or D) of the document (text) and the second row give the actual document (text).
I would like to count the number of occurence of words in all texts by categories (all text in category A, all text in category B).

I have tried to import an excel file. Column 1 (catégories) is set to "polynomial" and 'id". Column 2 (documents) is set to "text" and "attribute".

I connect the excel import box with the "exa" port of the ProcessDocumentFromData box. The program compiles well but gives no results : in the metadata view of the results window, it writes "example set : 0 examples, 0 special attributes, 0 regular attributes)...

What can I do?

Thanks,

Barthélémy