import nltk
from nltk.corpus import movie_reviews
from nltk.tokenize import word_tokenize
from nltk.stem import WordNetLemmatizer
from nltk.corpus import stopwords
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import accuracy_score, classification_report
import random

Step 2: Download required NLTK data

nltk.download(['movie_reviews', 'punkt', 'stopwords', 'wordnet', 'omw-1.4'])

Step 3: Initialize tools

lemmatizer = WordNetLemmatizer()
stop_words = set(stopwords.words('english'))
def preprocess_text(text):
    tokens = word_tokenize(text.lower())
    processed_tokens = [
        lemmatizer.lemmatize(token)
        for token in tokens
        if token.isalpha() and token not in stop_words
    ]
    return ' '.join(processed_tokens)

Step 4: Prepare balanced dataset

positive_ids = movie_reviews.fileids('pos')
negative_ids = movie_reviews.fileids('neg')
positive_reviews = [preprocess_text(' '.join(movie_reviews.words(fileid))) for fileid in positive_ids]
negative_reviews = [preprocess_text(' '.join(movie_reviews.words(fileid))) for fileid in negative_ids]

Step 5: Combine and label (1 for positive, 0 for negative)

all_reviews = positive_reviews + negative_reviews
labels = [1]*len(positive_reviews) + [0]*len(negative_reviews)

Step 6: Shuffle the data

combined = list(zip(all_reviews, labels))
random.shuffle(combined)
all_reviews, labels = zip(*combined)

Step 7: Use TF-IDF instead of simple Bag of Words

vectorizer = TfidfVectorizer(max_features=2000)
X = vectorizer.fit_transform(all_reviews)
y = labels

Step 8: Split data properly (80% train, 20% test)

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

Step 9: Train classifier

classifier = MultinomialNB()
classifier.fit(X_train, y_train)

Step 10: Evaluate

y_pred = classifier.predict(X_test)
print(f"Accuracy: {accuracy_score(y_test, y_pred):.2f}")
print("\nClassification Report:")
print(classification_report(y_test, y_pred))

Step 11: Example predictions

test_samples = [
    "This movie was fantastic! The acting was great and the plot was engaging.",
    "The film was terrible. I hated every minute of it.",
    "It was okay, not great but not awful either."
]
for review in test_samples:
    processed = preprocess_text(review)
    vector = vectorizer.transform([processed])
    prediction = classifier.predict(vector)
    print(f"\nReview: {review}")
    print(f"Prediction: {'Positive' if prediction[0] == 1 else 'Negative'}")

Output