The EWN sense ("WORD_MEANING") IDs are not preserved by the conversion modules,
since the new WordNet entries are assigned synset offsets according to their
positions in the new files. However, knowing which synset offset corresponds
with which EWN sense ID is often important, particularly when working with
sense-annotated corpora which use the EWN IDs.
It would be helpful if the conversion module would also produce a file with
such a mapping. Attached is a patch which implements this functionality.
Original issue reported on code.google.com by tristan.miller@nothingisreal.com on 25 Apr 2012 at 11:56
Original issue reported on code.google.com by
tristan.miller@nothingisreal.com
on 25 Apr 2012 at 11:56Attachments: