Anaphora resolution example requested with python-nltk -
i have been looking howto nltk-drt , module definitions nltk.sem.drt having hard time trying understand how achieve basic functionality using these packages.
an example of task achieve following: given sentence
google 1 of largest tech companies. employees many people.
i able write small function parses sentence such this, , return sentence such
google 1 of largest tech companies. google employees many people.
a more realistic test case can like
tom, dick , harry invited party, declined. made sherry angry.
being resolved to
tom, dick , harry invited sherry party, declined. made sherry angry.
i don't think drt (discourse representation theory) deals anaphora resolution, wanted. deals representing meaning of sentence in formal logic.
also, there name "more realistic test case", called "cataphora".
anaphora resolution difficult, can try stanford dcoref
and see examples here: anaphora resolution using stanford coref
Comments
Post a Comment