# RTextTools for coded sentences

I'm new to text mining and I'm not sure if could be applied here:

I have labeled sentences but the words I work with are codes, so each word length ranges from 5 to 15 letters and there are only 4 different letters (A, B, C, D).

Looks like this :

Sentence1: AAABCD BAAAA ABCCBAA CCCCC...... AAAAAAABB CCCCCCCAAA (Positive labeled) ...
Sentence10000: ACCCCCCCCCCCD ... AACCDD (Negative labeled)


Each sentence could be composed of a different number of words in a different order, ranging from 6 to 50 words.

Could RTextTools work in this case? When the matrix is created you have to select a language and I have no language.

-
This has been posted on SO and removed, which suggests you are looking for a statistical solution to your problem. It sounds, however, that some pieces of information are missing: What are you trying to do precisely? –  chl Sep 2 '12 at 11:51
I'm trying to find out if my sentences can be classified using RTextTools because my sentences are not in any language, they are just 4 letter codes. In the breast cancer example they classify numbers using RTextTools which makes me think that it might be used for my sentences –  Karen Sep 2 '12 at 12:58
Thanks for the information. What do these codes represent, and why are they of different length? –  chl Sep 3 '12 at 7:32