Fake News Detection System Using Django and Machine Learning

• Django admin panel for staff and superusers
• Manage all prediction results across all users
• Search, filter, edit, and delete detection records
• Filter records by user, input type, prediction result, and date
• Staff dashboard with full cross-user visibility
• View all users’ history, reports, and dashboard statistics
• Filter reports by user ID and other criteria
• Delete any visible record from staff queryset
• Management commands for model training, demo data seeding, migrations, and superuser creation

• User registration with email and password
• Secure login and logout system
• Password reset with email support
• Fake news detection from pasted news text
• Fake news detection from article URL
• Hybrid prediction using Logistic Regression and Random Forest
• Confidence score and per-model probability display
• Cleaned text preview for transparency
• Optional country, India state, and news category selection
• Personal dashboard with prediction history
• Search and filter previous detection records
• Delete own prediction records
• Reports page with charts and analytics
• Static About and Contact pages
• Health check endpoint for application readiness

• Hybrid machine learning pipeline with ensemble probability averaging
• TF-IDF text vectorization for feature extraction
• NLP preprocessing using NLTK stopwords and lemmatization
• URL article extraction using requests, BeautifulSoup, and lxml
• URL authenticity heuristics for additional analysis
• Country hint detection from URLs for reporting
• Model artifact saving using joblib
• Dashboard charts using matplotlib and seaborn
• CSV-based training support for custom datasets
• Dummy data generation using seed command
• Environment variable support with python-dotenv
• User-wise access control for privacy and security
• Staff-level reporting and management system
• Ready for local deployment and further production setup

• Create a virtual environment
  python -m venv .venv
• Activate the virtual environment
  Windows: .venv\Scripts\activate
  Linux/macOS: source .venv/bin/activate
• Install dependencies
  pip install -r requirements.txt
• Apply migrations
  python manage.py migrate
• Train the machine learning model
  Demo dataset:
  python manage.py train_model --csv data/sample_news.csv
  Full dataset:
  python manage.py train_model
• Optional: seed demo data
  python manage.py seed_data
  or
  python manage.py seed_data --user YOUR_USERNAME
• Create admin account
  python manage.py createsuperuser
• Run the development server
  python manage.py runserver
• Open in browser
  Home: http://127.0.0.1:8000/
  Register: http://127.0.0.1:8000/register/
  Login: http://127.0.0.1:8000/login/
  Dashboard: http://127.0.0.1:8000/dashboard/
  Admin: http://127.0.0.1:8000/admin/

This project does not include default login credentials.

• Create admin account using:
  python manage.py createsuperuser
• Normal users can register from:
  /register/