User-generated content can boost engagement, but it also carries risks: spam, hate speech, harassment, misinformation. A moderation system helps strike a balance: protecting users while preserving freedom of expression.
CensorPRO enables platforms to automatically and/or manually moderate user-generated content (text, images). Use it to filter profanity, hate speech, spam, or other undesired content and maintain a safe community.
- Automatic moderation: AI-based content moderation for Text and Images
- Manual moderation UI: Dashboard for moderators to review flagged content
- User management / roles (e.g. user, moderator, admin)
- API endpoints for frontend / external integration
- Backend/ — contains server-side code (Express, models, controllers, routes)
- Frontend/ — contains client-side application
- Root-level
package*.jsonmanages some shared dependencies or scripts
- Node.js (v14+ or your target version)
- (Optional) Database (MongoDB / Postgres / etc.)
- (Optional) Environment variables / API keys for moderation models
-
Clone the repo
git clone https://github.com/Nayanpawar03/CensorPRO.git cd CensorPRO -
Install dependencies for backend and frontend
cd Backend
npm install
cd ../Frontend
npm install- Setup environment variables. Create a .env file in Backend with content like:
PORT=5000
DB_URI=your_database_connection_string
MOD_MODEL_KEY=your_model_api_key- Run the app
cd..
npm run devUser posts content → frontend sends it to backend API
Backend moderation: checks the content via AI model (User gets ption to choose AI moderation or Admin review for moderation)
Shows result whether safe or unsafe
Moderator dashboard: view flagged items, approve or reject.
Backend: Node.js, Express
Frontend: React
Database: PostgreSQL
Moderation approach: ML model (DistillBert for Text moderation), external API (SightEngine for image moderation)