Skip to content

AOSSIE-Org/Social-Street-Smart

Repository files navigation

Social Street Smart

Social Street Smart is a Chrome extension aimed at making the internet a safer and more productive space for users. It addresses issues like abusive language, fake news, clickbait, malicious websites, and security attacks.

Project Overview

  • Frontend: Chrome extension built with React and TypeScript
  • Backend: Multiple Python-based APIs for various detection tasks
  • ML Models: Pre-trained models for clickbait, hate speech, and fake news detection

Features

  • Clickbait detection
  • Hate speech detection
  • Fake news detection
  • Disinformation in images detection
  • Web activity tracking
  • Website reputation checking

Quick Start

Frontend (Chrome Extension)

cd client
npm install
npm run build

Load the 'dist' folder as an unpacked extension in Chrome

Backend Servers

cd server
docker compose up

Project Structure

Social-Street-Smart/
├── client/                 # Frontend (Chrome extension)
├── server/                 # Backend services
│   ├── clickbait/
│   ├── hate-speech/
│   ├── fake-news/
│   ├── image-api/
│   └── news-origin/
├── ML/              # Machine learning models
│   ├── clickbait/
│   ├── hate-speech/
│   └── fake-news/
└── docker-compose.yml

API Endpoints

  • Clickbait API: http://localhost:5000/predict
  • Hate Speech API: http://localhost:5001/predict
  • Fake News API: http://localhost:5002/predict
  • Image Disinformation API: http://localhost:5003/analyze
  • News Origin API: http://localhost:5004/origin

Contributing

We welcome contributions to Social Street Smart! For detailed setup instructions and how to contribute, please see CONTRIBUTING.md.

License

This project is licensed under the CC-By-NC-ND 4.0 License - see the LICENSE file for details.

License

Acknowledgements

  • AOSSIE for organizing and supporting this project
  • All contributors and mentors who have helped shape Social Street Smart