🔖 Fake News Prediction Model

Hey, I created a Fake News Prediction Model using Logistic Regression and various text preprocessing techniques like stemming, stopword removal, and TF-IDF vectorization.

---

🚀 Project Overview

This project is machine learning model that classifies news articles as Real (1) or Fake (0) based on their textual content.

---

🔧 Technologies Used

• Python 3
• NumPy
• Pandas
• NLTK (Natural Language Toolkit)
• Scikit-learn (sklearn)

---

🛠️ Machine Learning Pipeline

1. Data Cleaning:

   • Lowercasing
   • Removing special characters & numbers
   • Removing stopwords
   • Stemming words (Porter Stemmer)

2. Feature Extraction:

   • Used TF-IDF Vectorizer to convert text into numeric features.

3. Model Training:

   • Used Logistic Regression for binary classification.

4. Model Evaluation:

   • Calculated Accuracy Score on the test set.

---

📁 Dataset

The dataset contains columns like:

• title
• text
• label → 0 for Fake, 1 for Real

---

🏁 How to Run

1. Install dependencies:

   pip install numpy pandas scikit-learn nltk

2. Run the Python script / notebook (FakeNews.ipynb).

---

📊 Example Accuracy

✅ Model Accuracy: Around 95.28%

---

🤖 Future Improvements

• Try other models like SVM, Random Forest.
• Tune hyperparameters for better accuracy.
• Add API endpoints using Flask/FastAPI.

---

📌 Author

Rakshit Jain (@rakshitjain23)