自然语言处理 |拆分和合并块

SplitRule 类：它根据指定的分割模式分割一个块。它被指定为 }{<.*> 即两个相对的花括号，由两边的图案包围。

MergeRule 类：它基于第一个块的结束和第二个块的开始将两个块合并在一起。它被指定为 {}<.*> 即大括号彼此面对。

如何执行这些步骤的示例

• 从句子树开始。
  

• 分块完成句子。
  

• 块被分成多个块。
  

• 带有确定符的块被分成单独的块。
  

• 以名词结尾的块与下一个块合并。
  

代码 #1 – 构建树

from nltk.chunk import RegexpParser
chunker = RegexpParser(r'''
NP:
{
<.*>*}
}{<.*>
<.*>}{

{}
''')
sent = [('the', 'DT'), ('sushi', 'NN'), ('roll', 'NN'), ('was', 'VBD'), 
        ('filled', 'VBN'), ('with', 'IN'), ('the', 'DT'), ('fish', 'NN')]
chunker.parse(sent)

输出：

Tree('S', [Tree('NP', [('the', 'DT'), ('sushi', 'NN'), ('roll', 'NN')]), 
Tree('NP', [('was', 'VBD'), ('filled', 'VBN'), ('with', 'IN')]), 
Tree('NP', [('the', 'DT'), ('fish', 'NN')])])

代码 #2 – 拆分和合并

# Loading Libraries
from nltk.chunk.regexp import ChunkString, ChunkRule, ChinkRule
from nltk.tree import Tree
from nltk.chunk.regexp import MergeRule, SplitRule
  
# Chunk String
chunk_string = ChunkString(Tree('S', sent))
print ("Chunk String : ", chunk_string)
  
# Applying Chunk Rule
ur = ChunkRule('
<.*>*', 'chunk determiner to noun')
ur.apply(chunk_string)
print ("\nApplied ChunkRule : ", chunk_string)
  
# Splitting
sr1 = SplitRule('', '<.*>', 'split after noun')
sr1.apply(chunk_string)
print ("\nSplitting Chunk String : ", chunk_string)
  
  
sr2 = SplitRule('<.*>', '
', 'split before determiner')
sr2.apply(chunk_string)
print ("\nFurther Splitting Chunk String : ", chunk_string)
  
# Merging
mr = MergeRule('', '', 'merge nouns')
mr.apply(chunk_string)
print ("\nMerging Chunk String : ", chunk_string)
  
# Back to Tree
chunk_string.to_chunkstruct()

输出：

Chunk String :   
            
   

Applied ChunkRule :  {
            
  }

Splitting Chunk String :  {
  }{}{      
  }

Further Splitting Chunk String :  {
  }{}{    }{
  }

Merging Chunk String :  {
    }{    }{
  }

Tree('S', [Tree('CHUNK', [('the', 'DT'), ('sushi', 'NN'), ('roll', 'NN')]), 
          Tree('CHUNK', [('was', 'VBD'), ('filled', 'VBN'), ('with', 'IN')]), 
          Tree('CHUNK', [('the', 'DT'), ('fish', 'NN')])])