🏆 Hackathon Project: An intelligent university matching system that uses AI to crawl university websites in real-time, analyze admission requirements, and provide personalized university recommendations with explainable scoring.

CampusCompass revolutionizes the college application process by combining real-time web crawling, AI-powered content analysis, and intelligent matching algorithms to help students find their perfect university fit. Unlike static databases, our system crawls university websites live to ensure the most up-to-date information.

• 🤖 AI-Powered Web Crawling: Uses Ollama LLM to intelligently analyze university web pages and extract relevant admission information
• 📊 Real-Time Analysis: Crawls university websites on-demand to get the latest requirements and program information
• 🎯 Smart Matching Algorithm: Multi-factor scoring system considering academic fit, research alignment, and program availability
• 📈 Live Progress Tracking: Real-time updates during the matching process with detailed progress indicators
• 📄 Comprehensive Reports: Generates detailed PDF and CSV reports with explanations for each match

┌─────────────────┐    ┌──────────────────┐    ┌─────────────────┐
│   React Frontend │    │  Django Backend  │    │  Ollama AI      │
│   (Material UI)  │◄──►│  (REST API)      │◄──►│  (Local LLM)    │
└─────────────────┘    └──────────────────┘    └─────────────────┘
                              │
                              ▼
                    ┌──────────────────┐
                    │  Enhanced        │
                    │  Web Crawler     │
                    │  (Multi-threaded)│
                    └──────────────────┘
                              │
                              ▼
                    ┌──────────────────┐
                    │  University      │
                    │  Websites        │
                    │  (Live Data)     │
                    └──────────────────┘

• Smart Profile Creation: Upload resume and cover letter for enhanced matching
• Personalized Matching: Get recommendations based on your academic background, interests, and goals
• Real-Time Progress: Watch as the system analyzes universities in real-time
• Detailed Explanations: Understand why each university was recommended
• Export Options: Download results as PDF reports or CSV files

• Advanced Crawling Engine: BFS-based crawler with intelligent link prioritization
• AI Content Analysis: LLM-powered extraction of admission requirements and program details
• Scalable Architecture: Designed to handle multiple concurrent matching requests
• Extensible Scoring: Modular scoring system that can be easily customized
• Rich API: RESTful API for integration with other systems

• Python 3.8+
• Node.js 16+
• Ollama (recommended for AI features)

# Clone and navigate to project
cd server

# Create virtual environment
python3 -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Setup database
python manage.py migrate

# Start Django server
python manage.py runserver 0.0.0.0:8000

# Navigate to frontend directory
cd frontend

# Install dependencies
npm install

# Start development server
npm run dev

# Open browser to http://localhost:5173

# Install Ollama from https://ollama.com/
# Pull a model (we recommend gpt-oss for best results)
ollama pull gpt-oss

# Ollama will automatically start on http://localhost:11434

Students create detailed profiles including:

• Academic background (GPA, test scores, degree level)
• Research interests and career goals
• Resume and cover letter upload
• Funding requirements and preferences

Our enhanced crawler:

• Performs BFS traversal of university domains
• Prioritizes admission and faculty pages
• Uses AI to identify relevant content
• Extracts key information like requirements, deadlines, and program details

The system uses Ollama LLM to:

• Analyze page content for relevance
• Extract admission requirements
• Identify faculty research areas
• Score content based on student profile

Multi-factor scoring considers:

• Academic Fit: GPA requirements, test scores, degree level match
• Research Alignment: Overlap between student interests and faculty research
• Program Availability: Relevant programs and specializations
• Content Quality: Depth and recency of information found

Generate detailed reports with:

• Ranked university list with scores
• Explanation for each recommendation
• Key requirements and deadlines
• Faculty and research highlights

• POST /api/profiles - Create/update student profile
• GET /api/profiles/{id} - Get profile details
• POST /api/profiles/{id}/resume - Upload resume
• POST /api/profiles/{id}/cover_letter - Upload cover letter

• GET /api/universities - List available universities
• POST /api/match - Start matching process
• GET /api/match/{run_id} - Get matching status and results
• GET /api/match/{run_id}/download/pdf - Download PDF report
• GET /api/match/{run_id}/download/csv - Download CSV results

# Ollama Configuration
OLLAMA_URL=http://localhost:11434  # Default Ollama endpoint
OLLAMA_MODEL=gpt-oss              # AI model to use

# Crawling Settings
MAX_CRAWL_PAGES=100               # Maximum pages per university
CRAWL_TIMEOUT=5                   # Request timeout in seconds
CRAWL_DEPTH=3                     # Maximum crawl depth

{
  "mode": "limited",        // "limited" or "all"
  "max_links": 100,        // Maximum links to crawl
  "timeout": 5,            // Request timeout
  "depth": 3               // Maximum depth
}

The system integrates with Ollama for advanced AI capabilities:

• Content Relevance Scoring: AI evaluates how relevant each page is to the student's profile
• Information Extraction: Automatically extracts admission requirements, deadlines, and program details
• Faculty Research Analysis: Identifies faculty members and their research areas
• Intelligent Link Following: AI decides which links are most likely to contain useful information

If Ollama is not available, the system gracefully falls back to keyword-based analysis.

Our sophisticated scoring system uses multiple factors:

# Simplified scoring formula
university_score = (
    academic_fit_score * 0.4 +
    research_alignment_score * 0.3 +
    content_quality_score * 0.2 +
    program_availability_score * 0.1
)

• Academic Fit: Compares student credentials with university requirements
• Research Alignment: Matches student interests with faculty research
• Content Quality: Rewards universities with comprehensive, up-to-date information
• Program Availability: Considers availability of relevant programs and specializations

UniversityMatch/
├── frontend/                 # React frontend
│   ├── src/
│   │   ├── components/      # React components
│   │   ├── context/         # React context providers
│   │   └── assets/          # Static assets
├── server/                  # Django backend
│   ├── matcher/             # Main application
│   │   ├── models.py        # Database models
│   │   ├── views.py         # API endpoints
│   │   ├── services/        # Business logic
│   │   │   ├── enhanced_crawler.py
│   │   │   ├── ollama_client.py
│   │   │   └── enhanced_university_matcher.py
│   └── student_profiles/    # Generated profiles
└── requirements.txt         # Python dependencies

• Backend: Django 4.2+, Django REST Framework
• Frontend: React 18, Material-UI, Vite
• AI: Ollama, Local LLM integration
• Web Crawling: BeautifulSoup, Requests
• Database: SQLite (development), PostgreSQL (production)

1. Real-Time Web Crawling: Unlike competitors using static data, we crawl live
2. AI-Powered Analysis: Uses local LLM for intelligent content understanding
3. Explainable Recommendations: Every match comes with detailed explanations
4. Comprehensive Profiling: Supports resume/cover letter analysis
5. Live Progress Tracking: Real-time updates during the matching process

• Multi-threaded crawling with intelligent prioritization
• Graceful fallback when AI services are unavailable
• Responsive real-time UI with progress indicators
• Comprehensive API with detailed documentation
• Scalable architecture ready for production deployment

• Resume Parsing: Automatic extraction of skills and experience from resumes
• Faculty Matching: Direct matching with specific faculty members
• Application Tracking: Integration with application deadlines and requirements
• Social Features: Student community and peer recommendations
• Mobile App: Native mobile applications for iOS and Android

We welcome contributions! This project was built for a hackathon but is designed to be extensible and maintainable.

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

This project is licensed under the MIT License - see the LICENSE file for details.

• Ollama Team for providing excellent local LLM capabilities
• Django & React Communities for robust frameworks
• University Websites for providing the data that makes this possible

Built with ❤️ for students seeking their perfect university match

This project demonstrates the power of combining AI, web crawling, and intelligent algorithms to solve real-world problems in education.