从斯坦福分析器的无上下文短语结构输出中提取信息

Question

The Stanford Parser (http://nlp.stanford.edu/software/lex-parser.shtml) gives context-free phrase structure trees as following. What is the best way to extract things like all the Noun Phrases(NP) and Verb Phrases(NP) in the tree? Is there any Python (or Java) library that can allow me to read structures like these? Thank you.

(ROOT
  (S
    (S
      (NP
        (NP (DT The) (JJS strongest) (NN rain))
        (VP
          (ADVP (RB ever))
          (VBN recorded)
          (PP (IN in)
            (NP (NNP India)))))
      (VP
        (VP (VBD shut)
          (PRT (RP down))
          (NP
            (NP (DT the) (JJ financial) (NN hub))
            (PP (IN of)
              (NP (NNP Mumbai)))))
        (, ,)
        (VP (VBD snapped)
          (NP (NN communication) (NNS lines)))
        (, ,)
        (VP (VBD closed)
          (NP (NNS airports)))
        (CC and)
        (VP (VBD forced)
          (NP
            (NP (NNS thousands))
            (PP (IN of)
              (NP (NNS people))))
          (S
            (VP (TO to)
              (VP
                (VP (VB sleep)
                  (PP (IN in)
                    (NP (PRP$ their) (NNS offices))))
                (CC or)
                (VP (VB walk)
                  (NP (NN home))
                  (PP (IN during)
                    (NP (DT the) (NN night))))))))))
    (, ,)
    (NP (NNS officials))
    (VP (VBD said)
      (NP-TMP (NN today)))
    (. .)))

Answer 1

Check out the Natural Language Toolkit (NLTK) at nltk.org .

The toolkit is written in Python and provides code for reading precisely these kinds of trees (as well as lots of other stuff).

Alternatively, you could write your own recursive function for doing this. It would be pretty straightforward.

---

Just for fun: here's a super simple implementation of what you want:

def parse():
  itr = iter(filter(lambda x: x, re.split("\\s+", s.replace('(', ' ( ').replace(')', ' ) '))))

  def _parse():
    stuff = []
    for x in itr:
      if x == ')':
        return stuff
      elif x == '(':
        stuff.append(_parse())
      else:
        stuff.append(x)
    return stuff

  return _parse()[0]

def find(parsed, tag):
  if parsed[0] == tag:
    yield parsed
  for x in parsed[1:]:
    for y in find(x, tag):
      yield y

p = parse()
np = find(p, 'NP')
for x in np:
  print x

yields:

['NP', ['NP', ['DT', 'The'], ['JJS', 'strongest'], ['NN', 'rain']], ['VP', ['ADVP', ['RB', 'ever']], ['VBN', 'recorded'], ['PP', ['IN', 'in'], ['NP', ['NNP', 'India']]]]]
['NP', ['DT', 'The'], ['JJS', 'strongest'], ['NN', 'rain']]
['NP', ['NNP', 'India']]
['NP', ['NP', ['DT', 'the'], ['JJ', 'financial'], ['NN', 'hub']], ['PP', ['IN', 'of' ['NP', ['NNP', 'Mumbai']]]]
['NP', ['DT', 'the'], ['JJ', 'financial'], ['NN', 'hub']]
['NP', ['NNP', 'Mumbai']]
['NP', ['NN', 'communication'], ['NNS', 'lines']]
['NP', ['NNS', 'airports']]
['NP', ['NP', ['NNS', 'thousands']], ['PP', ['IN', 'of'], ['NP', ['NNS', 'people']]]]
['NP', ['NNS', 'thousands']]
['NP', ['NNS', 'people']]
['NP', ['PRP$', 'their'], ['NNS', 'offices']]
['NP', ['NN', 'home']]
['NP', ['DT', 'the'], ['NN', 'night']]
['NP', ['NNS', 'officials']]