Chunks and Chinks

分块是根据单词的性质将相似单词分组在一起的过程。 在下面的示例中，我们定义了必须生成块的语法。 语法表示在创建块时将遵循的诸如名词和形容词等短语的序列。 块的图形输出如下所示。

import nltk
sentence = [("The", "DT"), ("small", "JJ"), ("red", "JJ"),("flower", "NN"), 
("flew", "VBD"), ("through", "IN"),  ("the", "DT"), ("window", "NN")]
grammar = "NP: {<nn>?<dt>*<jj>}" 
cp = nltk.RegexpParser(grammar)
result = cp.parse(sentence) 
print(result)
result.draw()
</jj></dt></nn>

改变语法，我们得到一个不同的输出，如下所示。

import nltk
sentence = [("The", "DT"), ("small", "JJ"), ("red", "JJ"),("flower", "NN"),
 ("flew", "VBD"), ("through", "IN"),  ("the", "DT"), ("window", "NN")]
grammar = "NP: {

Chinking

Chinking是从块中移除一系列令牌的过程。 如果令牌序列出现在块的中间，则删除这些令牌，留下两个已经存在的块。

import nltk
sentence = [("The", "DT"), ("small", "JJ"), ("red", "JJ"),("flower", "NN"), ("flew", "VBD"), ("through", "IN"),  ("the", "DT"), ("window", "NN")]
grammar = r"""
  NP:
    {<.*>+}         # Chunk everything
    }<jj>+{      # Chink sequences of JJ and NN
  """
chunkprofile = nltk.RegexpParser(grammar)
result = chunkprofile.parse(sentence) 
print(result)
result.draw()
</jj>

如您所见，符合语法标准的部分从名词短语中省略为单独的块。 提取不在所需块中的文本的过程称为 chinking。