使用带有NLTK |的块标签(而不是NER)在句子中创建关系 NLP

Question

我正在尝试创建自定义块标记并从中提取关系.以下是将我带到级联块树的代码.

grammar = r"""
  NPH: {<DT|JJ|NN.*>+}          # Chunk sequences of DT, JJ, NN
  PPH: {<IN><NP>}               # Chunk prepositions followed by NP
  VPH: {<VB.*><NP|PP|CLAUSE>+$} # Chunk verbs and their arguments
  CLAUSE: {<NP><VP>}           # Chunk NP, VP
  """
cp = nltk.RegexpParser(grammar)
sentence = [("Mary", "NN"), ("saw", "VBD"), ("the", "DT"), ("cat", "NN"),
    ("sit", "VB"), ("on", "IN"), ("the", "DT"), ("mat", "NN")]


chunked = cp.parse(sentence)

输出 -

(S(NPH Mary/NN)锯/ VBD(NPH/DT cat/NN)坐/ VB on/IN(NPH/DT垫/ NN))

现在我尝试使用nltk.sem.extract_rels函数提取NPH标记值与其间的文本之间的关系,但它似乎仅适用于使用ne_chunk函数生成的命名实体.

IN = re.compile(r'.*\bon\b')
for rel in nltk.sem.extract_rels('NPH', 'NPH', chunked,corpus='ieer',pattern = IN):
        print(nltk.sem.rtuple(rel))

这会出现以下错误 -

ValueError:尚未识别您的主题类型的值:NPH

有没有一种简单的方法只使用块标签来创建关系,因为我真的不想重新训练NER模型来检测我的块标签作为相应的命名实体

谢谢!

Answer 1

1. extract_rels(doc)检查参数subjclass并且objclass是已知的NE标签,因此错误NPH.

2. 简单,特别的方法是重写自定义extract_rels函数(下面的示例).

   import nltk
   import re
   
   grammar = r"""
     NPH: {<DT|JJ|NN.*>+}          # Chunk sequences of DT, JJ, NN
     PPH: {<IN><NP>}               # Chunk prepositions followed by NP
     VPH: {<VB.*><NP|PP|CLAUSE>+$} # Chunk verbs and their arguments
     CLAUSE: {<NP><VP>}           # Chunk NP, VP
     """
   cp = nltk.RegexpParser(grammar)
   sentence = [("Mary", "NN"), ("saw", "VBD"), ("the", "DT"), ("cat", "NN"),
       ("sit", "VB"), ("on", "IN"), ("the", "DT"), ("mat", "NN")]
   
   chunked = cp.parse(sentence)
   
   IN = re.compile(r'.*\bon\b')
   
   def extract_rels(subjclass, objclass, chunked, pattern):
   
       # padding because this function checks right context
       pairs = nltk.sem.relextract.tree2semi_rel(chunked) + [[[]]] 
   
       reldicts = nltk.sem.relextract.semi_rel2reldict(pairs)
   
       relfilter = lambda x: (x['subjclass'] == subjclass and
                              pattern.match(x['filler']) and
                              x['objclass'] == objclass)
   
   
       return list(filter(relfilter, reldicts))
   
   for e in extract_rels('NPH', 'NPH', chunked, pattern=IN):
       print(nltk.sem.rtuple(e))
   

   输出:

   [NPH: 'the/DT cat/NN'] 'sit/VB on/IN' [NPH: 'the/DT mat/NN']