Fake News Detector

A machine learning-based fake news detection system built with Flask, PyTorch, and Transformers.

Author

Bishwaprotap Ray

• Role: Software Developer Intern | AI & Machine Learning Engineer
• Education: B.Sc. in Computer Science & Engineering (International University of Business Agriculture and Technology)
• Specialization: AI, Machine Learning, LLM, FastAPI, Voice Assistant Development
• Location: Dhaka, Bangladesh
• Mobile: +8801788974534
• Email: baburay214@gmail.com
• LinkedIn: https://www.linkedin.com/in/bishwaprotap-ray/
• GitHub: https://github.com/Bishwaprotapi

Technologies Used

• Python - Core programming language
• Flask - Web framework for the API
• PyTorch - Deep learning framework
• Transformers (Hugging Face) - Pre-trained NLP models
• Scikit-learn - Machine learning utilities
• Pandas - Data manipulation and analysis
• NumPy - Numerical computing
• BeautifulSoup4 - Web scraping
• Requests - HTTP library

Features

• Detect fake news using advanced NLP models
• Web-based interface for easy interaction
• RESTful API for integration
• Data scraping capabilities
• Model training and evaluation

Installation

1. Clone the repository:

git clone https://github.com/Bishwaprotapi/Fake_News_Detector.git
cd Fake_News_Detector

2. Create a virtual environment:

python -m venv .venv
.venv\Scripts\activate  # On Windows
source .venv/bin/activate  # On Linux/Mac

3. Install dependencies:

pip install -r requirements.txt

Usage

1. Run the Flask application:

python app.py

2. Open your browser and navigate to http://localhost:5000

3. Enter news text to check if it's fake or real

Project Structure

Fake_News_Detector/
├── app.py              # Main Flask application
├── load_dataset.py     # Dataset loading utilities
├── save_tokenizer.py   # Tokenizer saving utilities
├── model/              # Model files and prediction logic
├── data/               # Dataset files
├── scraper/            # Web scraping modules
├── templates/          # HTML templates
├── utils/              # Utility functions
└── requirements.txt    # Python dependencies

API Endpoint

• POST /predict - Submit news text for fake news detection
  • Request body: {"text": "your news text here"}
  • Response: {"result": "fake" or "real"}

License

This project is open source and available under the MIT License.