NLP Exp 5

import nltk
from nltk import ngrams
from nltk.tokenize import word_tokenize
from collections import Counter
#nltk.download('punkt')

text = """
Natural Language Processing with Python and NLTK is powerful.
It helps in building models to understand and generate human language.
"""
tokens = word_tokenize(text.lower())

def generate_ngrams(tokens,n):
    n_grams = list(ngrams(tokens,n))
    return n_grams

bigrams = generate_ngrams(tokens,2)
trigrams = generate_ngrams(tokens,3)

bigram_freq = Counter(bigrams)
trigram_freq = Counter(trigrams)

print("Bigrams:")
for gram,freq in bigram_freq.items():
    print(f"{gram}: {freq}")

print("\nTrigrams:")
for gram,freq in trigram_freq.items():
    print(f"{gram}: {freq}")



def predict_next_word(input_word, bigram_freq):
    candidates = {second: freq for (first, second), freq in bigram_freq.items() if first == input_word.lower()}
    if candidates:
        predicted = max(candidates, key=candidates.get)
        return predicted
    else:
        return None

input_word = input("Enter word: ")
next_word = predict_next_word(input_word, bigram_freq)

if next_word:
    print(f"Predicted next word after '{input_word}' is: {next_word}")
else:
    print(f"No prediction available for '{input_word}'")


def predict_next_word_trigram(word1, word2, trigram_freq):
    candidates = {
        third: freq
        for (first, second, third), freq in trigram_freq.items()
        if first == word1.lower() and second == word2.lower()
    }
    if candidates:
        predicted = max(candidates, key=candidates.get)
        return predicted
    else:
        return None

input_word1 = input("Enter word 1: ")
input_word2 = input("Enter word 2: ")
next_word = predict_next_word_trigram(input_word1, input_word2, trigram_freq)

if next_word:
    print(f"Predicted next word after '{input_word1} {input_word2}' is: {next_word}")
else:
    print(f"No prediction available for '{input_word1} {input_word2}'")