Search code examples
pythonnlpnltk

Python, NLP: How to find all trigrams from text files with adjectives as the middle term

I think the question is self-explanatory but here goes the detailed meaning of the question.

I want to extract all trigrams from text files using the nltk library having adjectives as the middle term.

Example Text - A red ball was with the good boy.

Example of output -

('A','red','ball'), ('the','good','boy')

and so on

Solution

  • This code should do it:

    import nltk
from nltk.tokenize import word_tokenize

nltk.download('punkt')
nltk.download('averaged_perceptron_tagger')

text = word_tokenize("He is a very handsome man. Her childern are funny. She has a lovely voice")
text_tags = nltk.pos_tag(text)
results = list()
for i, (txt, tag) in enumerate(text_tags):
    if tag in ["JJ", "JJR", "JJS"]:
        if (i > 0) and (i < len(text_tags)-1):
            results.append((text_tags[i-1][0], txt, text_tags[i+1][0]))

# output: [('very', 'handsome', 'man'), ('are', 'funny', '.'), ('a', 'lovely', 'voice')]