📰 Fake News Detection using Naïve Bayes

📌 Project Overview

This project builds a Fake News Detector using Natural Language Processing (NLP) and Machine Learning. We use the Naïve Bayes Classifier trained on the Fake.csv dataset to classify news articles as FAKE or REAL.

---

🚀 Installation

Ensure you have Python installed, then install the required dependencies:

pip install pandas numpy matplotlib seaborn scikit-learn

---

📂 Project Structure

📁 fake-news-detection
│── 📄 fake_news_detection.py   # Main script
│── 📄 README.md                # Project documentation
│── 📊 Fake.csv                  # Dataset (should be placed here)

---

🛠️ Steps Involved

1️⃣ Import Necessary Libraries

import pandas as pd
import numpy as np
import re
import string
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix

---

2️⃣ Load the Dataset

df = pd.read_csv("Fake.csv", on_bad_lines="skip", encoding='utf-8')

print(df.columns)  # Display column names
print(df['subject'].unique())  # Check unique values in 'subject'
df.rename(columns={'subject': 'label'}, inplace=True)  # Rename 'subject' to 'label'

---

🧹 3️⃣ Data Cleaning

def clean_text(text):
    text = text.lower()  # Convert to lowercase
    text = re.sub(r'\[.*?\]', '', text)  # Remove text inside brackets
    text = re.sub(r"https?://\S+|www\.\S+", '', text)  # Remove URLs
    text = re.sub(r"<.*?>+", '', text)  # Remove HTML tags
    text = re.sub(r"[^\w\s]", '', text)  # Remove punctuation
    text = text.strip()  # Remove leading/trailing spaces
    return text

df['content'] = df['content'].apply(clean_text)

---

✂️ 4️⃣ Splitting the Data

X = df['content']
y = df['label']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

---

🔡 5️⃣ Convert Text to Numerical Features

vectorizer = TfidfVectorizer(max_features=5000)
X_train_tfidf = vectorizer.fit_transform(X_train)
X_test_tfidf = vectorizer.transform(X_test)

---

🔥 6️⃣ Train the Naïve Bayes Model

model = MultinomialNB()
model.fit(X_train_tfidf, y_train)

---

🔍 7️⃣ Making Predictions

y_pred = model.predict(X_test_tfidf)

---

📈 8️⃣ Model Evaluation

print(f"Accuracy: {accuracy_score(y_test, y_pred):.2f}")
print("\nClassification Report:\n", classification_report(y_test, y_pred))

# Confusion Matrix
plt.figure(figsize=(6,4))
sns.heatmap(confusion_matrix(y_test, y_pred), annot=True, fmt='d', cmap='Blues')
plt.xlabel('Predicted')
plt.ylabel('Actual')
plt.title('Confusion Matrix')
plt.show()

📌 Sample Confusion Matrix:

---

📰 9️⃣ Fake News Detection Function

def predict_news(news_text):
    news_text = clean_text(news_text)
    news_vectorized = vectorizer.transform([news_text])
    prediction = model.predict(news_vectorized)
    return "FAKE NEWS" if prediction[0] == 1 else "REAL NEWS"

# Example Usage
news = "Breaking news! The direct dbt of 72000 per month will be given to each person"
print("\nPrediction:", predict_news(news))

---

🎯 Key Findings

✅ Data preprocessing (cleaning text) improved model accuracy.
✅ TF-IDF vectorization helped in feature extraction.
✅ The Naïve Bayes model effectively classifies fake news.

---

📌 Future Improvements

🔹 Try different machine learning models (e.g., Random Forest, LSTM).
🔹 Improve text preprocessing by removing stopwords.
🔹 Deploy the model using Flask or Streamlit for a web-based interface.

---

💡 Contributors

👤 Your Name | GitHub Profile

⭐ If you like this project, don't forget to give it a star on GitHub! ⭐

---

🚀 Happy Coding & Stay Informed! 📰