@AlanoudF it depends which other languages are in your dataset. For languages with different alphabets (Cyrillic, Japanese etc.) you can simply flag those characters. If you want to distinguish languages that both use Latin characters, this will be a lot harder. You could look for characters that never appear in English (like è, ö etc.), or use a lookup table of English words to compare to. As @suby says, if you can provide some sample dataset we can give some better advice.
