Skip to content
Related Articles

Related Articles

NLP | Singularizing Plural Nouns and Swapping Infinite Phrases
  • Last Updated : 26 Feb, 2019

Let’s understand this with an example :

  1. Is our child training enough?
  2. Is are our child training enough?

The verb ‘is’ can only be used with singular nouns. For plural nouns, we use ‘are’. This problem is very common in the real world and we can correct this mistake by creating verb correction mappings that are used depending on whether there’s plural or singular noun in the chunk.

Code #1 : singularize_plural_noun() class




def singularize_plural_noun(chunk):
      
    nnsidx = first_chunk_index(chunk, tag_equals('NNS'))
      
    if nnsidx is not None and 
    nnsidx + 1 < len(chunk) and 
    chunk[nnsidx + 1][1][:2] == 'NN':
          
        noun, nnstag = chunk[nnsidx]
        chunk[nnsidx] = (noun.rstrip('s'), nnstag.rstrip('S'))
          
    return chunk

 
Code #2 : Singularizing Plurals




singularize_plural_noun([('recipes', 'NNS'), ('book', 'NN')])

Output :



[('recipe', 'NN'), ('book', 'NN')]

The code above looks for the tag NNS to look for Plural Noun. After founding it, if the next word is a noun (determined by making sure the tag starts with NN), then we depluralize the plural noun by removing ‘s’ from the right side of both the tag and the word.

Swapping Infinite Phrases
An infinitive phrase is of the form A of B, such as ‘world of movies’. On can transform this to ‘movies world’ and it still holds the same meaning.

Code #3 : Let’s understand the swap_infinitive_phrase() class




def swap_infinitive_phrase(chunk):
    def inpred(wt):
        word, tag = wt
        return tag == 'IN' and word != 'like'
    
    inidx = first_chunk_index(chunk, inpred)
      
    if inidx is None:
        return chunk
      
    nnidx = first_chunk_index(chunk, 
                              tag_startswith('NN'), 
                              start = inidx, step =-1) or 0
                                
    return chunk[:nnidx] + chunk[inidx + 1:] + chunk[nnidx:inidx]

 
Code #4 : Let’s evaluate swap_infinitive_phrase




from transforms import swap_infinitive_phrase
  
swap_infinitive_phrase([('book', 'NN'), 
                        ('of', 'IN'), ('recipes', 'NNS')])

Output :

[('recipes', 'NNS'), ('book', 'NN')]

machine-learning

My Personal Notes arrow_drop_up
Recommended Articles
Page :