algorithm - Rewriting sentences while retaining semantic meaning -


is possible use wordnet rewrite sentence semantic meaning of sentence still ways same (or same)?

let's have sentence:

obama met putin last week. 
  1. is possible use wordnet rephrase sentence alternatives like:

    obama , putin met previous week. obama , putin met each other week ago. 
  2. if changing sentence structure not possible, can wordnet used replace relevant synonyms?

    for example:

    obama met putin previous week. 

if question possibility use wordnet sentence paraphrases. possible grammatical/syntax components. need system that:

  • first individual semantics of tokens , parse sentence syntax.
  • then understand overall semantics of composite sentence (especially if it's metaphorical)
  • then rehash sentence grammatical generator.

up till know of ace parser/generator can takes lot of hacking system make work paraphrase generator. http://sweaglesw.org/linguistics/ace/

so answer questions,

is possible use wordnet rephrase sentence alternatives? sadly, wordnet isn't silverbullet. need more semantics paraphrase task.

if changing sentence structure not possible, can wordnet used replace relevant synonyms? yes possible. figure out synonym replace-able hard... , need morphology/syntax component.

first run problem of multiple senses per word:

from nltk.corpus import wordnet wn sent = "obama met putin previous week"  in sent.split():     possible_senses = wn.synsets(i)     print i, len(possible_senses), possible_senses 

[out]:

obama 0 [] met 13 [synset('meet.v.01'), synset('meet.v.02'), synset('converge.v.01'), synset('meet.v.04'), synset('meet.v.05'), synset('meet.v.06'), synset('meet.v.07'), synset('meet.v.08'), synset('meet.v.09'), synset('meet.v.10'), synset('meet.v.11'), synset('suffer.v.10'), synset('touch.v.05')] putin 1 [synset('putin.n.01')] 0 [] previous 3 [synset('previous.s.01'), synset('former.s.03'), synset('previous.s.03')] week 3 [synset('week.n.01'), synset('workweek.n.01'), synset('week.n.03')] 

then if know sense (let's first sense), multiple words per sense , not every word can replaced in sentence. moreover, in lemma form not surface form (e.g. verbs in base form (simple present tense) , nouns in singular):

from nltk.corpus import wordnet wn sent = "obama met putin previous week"  in sent.split():     possible_senses = wn.synsets(i)     if possible_senses:         print i, possible_senses[0].lemma_names     else:         print 

[out]:

obama met ['meet', 'run_into', 'encounter', 'run_across', 'come_across', 'see'] putin ['putin', 'vladimir_putin', 'vladimir_vladimirovich_putin'] previous ['previous', 'old'] week ['week', 'hebdomad'] 

Comments

Popular posts from this blog

PHPMotion implementation - URL based videos (Hosted on separate location) -

c# - Unity IoC Lifetime per HttpRequest for UserStore -

I am trying to solve the error message 'incompatible ranks 0 and 1 in assignment' in a fortran 95 program. -