Notes

import re
from nltk import bigrams, FreqDist
from nltk.corpus import stopwords
from nltk.tokenize import regexp_tokenize
from nltk.collocations import BigramCollocationFinder
from nltk.metrics import BigramAssocMeasures

def performBigramsAndCollocations(textcontent, word):
# Step 1: Tokenize words using regex tokenizer
tokenizedwords = regexp_tokenize(textcontent, pattern=r'w+')

# Step 2: Convert all tokens to lowercase
tokenizedwords = [token.lower() for token in tokenizedwords]

# Step 3: Compute bigrams
tokenizedwordsbigrams = list(bigrams(tokenizedwords))

# Step 4: Filter out bigrams containing stopwords
stop_words = set(stopwords.words('english'))
tokenizednonstopwordsbigrams = [
bigram for bigram in tokenizedwordsbigrams
if bigram[0] not in stop_words and bigram[1] not in stop_words
]

# Step 5: Compute frequency of filtered bigrams
bigram_freq = FreqDist(tokenizednonstopwordsbigrams)

# Step 6: Determine most frequent word after the given word
mostfrequentwordafter = None
following_words = [b[1] for b in tokenizedwordsbigrams if b[0] == word]
if following_words:
mostfrequentwordafter = FreqDist(following_words).max()

# Step 7: Generate collocations using BigramCollocationFinder
bigram_finder = BigramCollocationFinder.from_words(tokenizedwords)
bigram_finder.apply_word_filter(lambda w: w in stop_words)
collocationwords = bigram_finder.nbest(BigramAssocMeasures.likelihood_ratio, 10)

return mostfrequentwordafter, collocationwords

if __name__ == "__main__":
# Example usage
textcontent = "This is a sample text to demonstrate bigram and collocation extraction using Python."
word = "to"
print(performBigramsAndCollocations(textcontent, word))