text classification with dialect in Arabic language

Badr

I made text classification for Arabic and it works very well with standard Arabic language but now I will use   with dataset with different dialect in Arabic ? can I use same operators( tokenize and stem(Arabic)  and stopword 

Telcontar120

It should as long as the underlying characters are not different, even though the vocabulary, syntax and usage may vary in the dialect.  RapidMiner (or any NLP algorithm) doesn't really understand languages, it just transforms them all into numerical representations to manipulate.