I am working on a corpus of learner English which is annotated with
grammatical tags. I would like to achieve two tasks:
-retrieve the error tags
-retrieve the corrected form in alphabetical which I will be in the corpus
(next to error)
Sharon Denness