# Program 7
# # Use a dataset of customer product reviews (e.g., Amazon reviews) to classify the sentiment
# of each review as positive, negative, or neutral using a pre-trained machine learning
# model (e.g., Naïve Bayes). Evaluate the accuracy of your sentiment classifier.
# Dataset Link: https://www.kaggle.com/datasets/snap/amazon-fine-food-reviews

import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split, cross_val_score
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix

# Step 1: Load the dataset
df = pd.read_csv(r"<Path>")
df = df[["Text", "Score"]].dropna()


# Step 2: Map Score to Sentiment (3-class)
def map_sentiment(score):
    if score <= 2:
        return -1  # Negative
    elif score == 3:
        return 0  # Neutral
    else:
        return 1  # Positive


df["Sentiment"] = df["Score"].apply(map_sentiment)
# Step 3: Limit sample size to 20,000 (unbalanced)
df_sampled = df.sample(n=20000, random_state=42)
# Step 4: Vectorize text using CountVectorizer
vectorizer = CountVectorizer(stop_words="english", max_features=5000)
X = vectorizer.fit_transform(df_sampled["Text"])
y = df_sampled["Sentiment"]
# Step 5: Train/test split
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.3, random_state=42
)
# Step 6: Train Naive Bayes classifier
model = MultinomialNB()
model.fit(X_train, y_train)
# Step 7: Predict
y_pred = model.predict(X_test)
# Step 8: Evaluation
accuracy = accuracy_score(y_test, y_pred)
print(f"Accuracy: {accuracy * 100:.2f}%\n")
print("Classification Report:")
print(
    classification_report(
        y_test,
        y_pred,
        zero_division=1,
        target_names=["Negative", "Neutral", "Positive"],
    )
)
print("Confusion Matrix:")
print(confusion_matrix(y_test, y_pred))
# Step 9: Cross-validation
cross_val_accuracy = cross_val_score(model, X, y, cv=5, scoring="accuracy")
print(f"\nCross-validation accuracy: {cross_val_accuracy.mean() * 100:.2f}%")
sentiment_counts = df_sampled["Sentiment"].value_counts()
sentiment_labels = [
    "Positive" if i == 1 else "Neutral" if i == 0 else "Negative"
    for i in sentiment_counts.index
]
# Plot a Pie Chart – Sentiment Proportion
plt.figure(figsize=(6, 6))
plt.pie(
    sentiment_counts,
    labels=sentiment_labels,
    autopct="%1.1f%%",
    colors=["green", "red", "orange"],
    startangle=140,
)
plt.title("Sentiment Proportion (Unbalanced)")
plt.tight_layout()
plt.show()