Peer-Review Platform Documentation
Welcome to Peer-Review Platform
The Peer-Review Platform is a comprehensive documentation analysis and improvement tool that combines traditional style checking with AI-powered content enhancement. This system analyzes documents for style compliance, detects ambiguities, and provides intelligent suggestions for improvement.
Key Features
-
Multi-format Support: Analyze PDF, DOCX, Markdown, and AsciiDoc documents
-
45+ Style Rules: Comprehensive rule system covering grammar, punctuation, structure, and formatting
-
AI-powered Rewriting: Two-pass improvement system using Ollama integration
-
Ambiguity Detection: Specialized detectors for fabrication risks, missing actors, and unclear references
-
Real-time Analysis: WebSocket-based communication for instant feedback
-
Extensible Architecture: Modular design for easy extension and customization
Getting Started
-
Architecture: System Architecture - Understanding the complete system design and component interactions.
-
How-to Guides: Step-by-step guides for extending the system:
-
API Reference: API Documentation - Complete reference for all API endpoints and WebSocket interfaces.
Technology Stack
-
Backend: Flask 3.0+, Python 3.8+
-
AI/ML: SpaCy 3.7+, Transformers 4.36+, Ollama 0.1+
-
Document Processing: PyPDF2, python-docx, python-markdown
-
Frontend: HTML5, JavaScript ES6+, WebSockets
-
Deployment: Docker, Docker Compose
Architecture Overview
The system follows a modular architecture with clear separation of concerns:
graph TD A[Document Input] --> B[Format Detection] B --> C[Document Parser] C --> D[Style Analyzer] D --> E[Rules Engine] D --> F[Ambiguity Detector] D --> G[AI Rewriter] E --> H[Analysis Results] F --> H G --> H H --> I[Client Interface]
Quick Start
-
Installation
git clone https://github.com/gtrivedi88/peer-lens.git cd peer-lens pip install -r requirements.txt
-
Configuration
cp config.example.py config.py # Edit config.py with your settings
-
Run the Application
python app.py
-
Access the Interface
Open your browser to
http://localhost:5000
Support
For questions, issues, or contributions:
-
Documentation: Browse the architecture and how-to guides
-
Issues: Report bugs or request features on GitHub
-
Contributions: See the contribution guidelines for development setup
License
This project is licensed under the MIT License - see the LICENSE file for details.