load text files

ghina84

Goodmorning everybody,

from the documentation I found on the website I cannot understand which operator I should use to load a serie of text files (.txt or .xml).

Can you help me please?

Thank you,

Laura

land

Hi Laura,
surprisingly you should use an operator called "TextInput". You can specify directories, where the texts are read from, in the parameter texts. Each directory listed there is searched for text files and each text file becomes an example. A directory has to contain all examples of one label, since the directory structure is used for labeling the data.

Greetings,
  Sebastian

ghina84

surprisingly I already tried it...

but instead of gettin a matrix like this:

rows=documents
columns=terms

I get a matrix like this:

rows=id
columns=documents (i.e. each attribute is one ENTIRE document)

is it normal?...

DPierre

Where can I find the TextInput operator?

sgenzer

hi @DPierre this is an old thread. Try downloading the Text Processing extension from the marketplace and then using "Read Document". There is a good set of tutorials on the Academy for this: https://academy.rapidminer.com/courses/text-and-web-mining-with-rapidminer

Scott